对照上面的退款运费所有情形下都由卖家承担 可以设置自动批准退款

Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

Learn more at Review Commons

Reply to the reviewers

'The authors do not wish to provide a response at this time.'

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

Referee #3

Evidence, reproducibility and clarity

Liver cancer is a deadly disease that consists of hepatocellular carcinoma (HCC) and cholangiocarcinoma (CCA). CCA can originate from hepatocytes. Interestingly, the NRF2 pathway is hyperactivated in 15% of HCC. To explore this, the authors developed a model of DOX-inducible hyperactivated NRF2T80K in hepatocytes of zebrafish. They suggest that hepatocyte to cholangiocyte transdifferentiation occurs, resulting in expansion of the cholangiocyte compartment. Importantly, they explore these phenotypes in larvae and adult zebrafish, clearly demonstrating that the impact of NRF2-driven cell plasticity is not limited to conditions of organogenesis. Using a pharmacologic inhibitor of BRG1/BRM, they also suggest that NRF2-driven cell plasticity is dependent on the SWI/SNF complex

Major comments

A major question regards the involvement of the SWI/SNF complex. The authors state, "NRF2 co-opts the SWI/SNF complex to drive liver cell plasticity". However, these conclusions are largely based on the phenotypes using the BRG1 inhibitor FHD-286. Currently, it is unclear the degree of on-target activity FHD-286 has on BRG1, and it would be important to potentially explore this. Further, did any other BRG1 inhibitors score from the screen, or could they explore BRG1 inhibition using distinct pharmacologic or genetic approaches? These experiments would strengthen the conclusions of the study.

Significance

This is a very interesting paper with broad implications. In future studies, it would be interesting to explore how NRF2 potentially impacts the SWI/SNF complex. It would also be interesting to explore which target genes downstream of NRF2 hyperactivation drive expansion of the cholangiocyte compartment. Finally, it would be interesting (in future studies) to explore metabolic changes caused by NRF2 hyperactivation in hepatocytes of zebrafish, and how these changes could cause transdifferentiation of hepatocytes into cholangiocytes

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

Referee #2

Evidence, reproducibility and clarity

Summary:

In the present manuscript, Dr. Ong and colleagues investigated the role of NRF2 activation in promoting liver cell plasticity. To support their conclusions, the authors use models of constitutive NRF2 activation, including KEAP1 KO and expression of the mutated NRF2T80K in HepG2 cells. In addition, they developed a transgenic zebrafish model with hepatocyte-specific inducible expression of mutated NRF2T80K to investigate the effect of NRF2 activation in liver tumour initiation. They found out that NRF2T80K expression promotes expansion of cholangiocyte compartment that resulted from hepatocyte transdifferentiation and provide evidence suggesting the potential role of SWI/SNF chromatin remodeling complex in this process.

Major comments

1. The major concern is to rule out the possibility that the observed phenotypes are influenced by off-target effects of doxycycline treatment in zebrafish. Although some control experiments are presented, they were performed at different experimental settings, or the methodology is not sufficiently clear. Therefore, control conditions should be presented more clearly. In particular, the authors should include analyses of liver plasticity markers and NRF2 activity markers in wild-type control animals treated with doxycycline. This would help demonstrate that the observed effects are due to NRF2T80K activation rather than doxycycline treatment itself. In addition, the experimental design is unclear in several panels (e.g. It is not clear if the zebrafish images were taken at 10 dpf and whether that also means the time of NRF2T80K induction).
2. It would be important to determine whether systemic activation of NRF2WT on KEAP1 KO background produces similar effects.
3. The data shown with the model keap1a/keap1b crispants needs more detailed explanation.
4. The authors should better explain the criteria to select the hit from the compound screen. Does the selected compound also have effect on KEAP1 KO or NRF2 OE cells?
5. Please double check the indications of treatment durations (time points/days) throughout the main text, figure legends, and experimental schemes to ensure consistency. Including a schematic overview that clearly distinguishes the larval and adult experimental settings would improve clarity and help readers follow the experimental design.
6. The authors conclude the results section with "Together, these data highlight a role for SWI/SNF-dependent chromatin remodelling complex in regulating liver cell plasticity in the context of NRF2 pathway activation." However, the data presented only suggests a possible role of SWI/SNF. The conclusion is based solely on pharmacological treatment, without direct functional validation of the specific components (e.g. through genetic perturbations). Therefore, the current evidence supports an association rather than a definitive role in regulation of liver cell plasticity.
7. Main conclusions derived from the imaging and gene expression analysis should be further validated at the protein level. Confirming the observed changes in the protein expression would provide stronger support for the proposed biological mechanism.

Minor comments

1. Immunoblot of KEAP1 is missing to confirm KO in Fig1B and Immunoblot of NRF2 is missing to confirm OE in Fig1D.
2. Review whether the data normalization appropriate matches the statistical test applied (In FigS1A, are the reference/control values random set to 1 or the references values used were the average of the controls?).
3. Please provide a more detailed explanation of the data processing and curation used to generate the graphs on Fig4B.
4. The legend indication of the panels on FigS4 is wrong. Please revise all figure legends and main text.

Significance

The conclusions are relevant for the field of basic research on cancer and cell biology. However, would be essential to clarify the points raised. The major limitation is that conclusions were made based on one model.

Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

Learn more at Review Commons

Referee #1

Evidence, reproducibility and clarity

While the study is technically ambitious, challenging in vivo zebrafish, scRNA-seq, and chemical library screen. On the other hand, the NRF2-SWI/SNF causal link is not well supported yet to fully back the claims. This reviewer's background is in redox signaling, so does not comment on the zebrafish technical methodology.

Major comments

1. No wild-type NRF2 control is studied. Everything is done only with NRF2T80K, but no WT NRF2 overexpression comparison. Therefore, this reviewer cannot tell if the phenotype is mutant-specific or just a general consequence of NRF2 pathway activation.
2. FHD-286 does not establish an NRF2-SWI/SNF link. As BRG1/BRM are broadly acting, but not NRF2-specific, identification of FHD-286 does not prove SWI/SNF's action downstream of NRF2. In fact, Figs. S4C & D show that FHD-286 does not fully suppress NRF2 target genes even while blocking trans-differentiation. This disconnect suggests FHD-286 may act on a parallel or downstream pathway, not addressed in this version.
3. Several other hits in the Fig. 4A screen show similar suppression to FHD-286. The authors need to explain the rationale for picking FHD-286 as the lead compound.
4. There are no functional cancer assays. Only histology/fluorescence for trans-differentiation were conducted, but no clonogenic assay, anchorage-independent growth, organoids, nor transplant tumor-initiation studies. The link to actual tumorigenesis remains speculative.
5. The Introduction says NRF2-driven tumorigenesis mechanisms are "poorly understood", but there are substantial existing literatures (e.g., PMID: 38308097, 37364049, 31961719).
6. No mention about where NRF2T80K mutation was first identified, its prevalence across liver cancer subtypes, nor known functional consequences beyond disrupting KEAP1 binding. This point should be stated more in detail.
7. In Fig 2A, text describes CCA-like glandular structures after DOX treatment, but they are not marked with arrows or outlines in the figure.

Minor comments

1. GSEA results would benefit from heatmaps showing the leading-edge genes driving the enrichment scores.
2. NQO1 and HMOX1 are the standard markers as NRF2 target genes. Incorporating them would strengthen the validation of NRF2 pathway engagement in this model.
3. The figure legend for Figure 1G states that it shows magnifications from Figure 1G. This should refer to Figure 1E

Significance

This manuscript uses a zebrafish model with inducible NRF2T80K expression, plus HepG2 cells, to show that constitutive NRF2 activation drives hepatocyte-to-cholangiocyte trans-differentiation. This phenotype is cell-autonomous, conserved between zebrafish and human cells, and reversible. The authors conducted a small-molecule screen to identify the BRG1/BRM inhibitor FHD-286 as a suppressor of this trans-differentiation. The authors propose this screening result links NRF2 to the SWI/SNF complex. The phenotypic work is clear and interesting, but the mechanistic experimental design has fundamental gaps that remain unaddressed.

If the mechanism holds up, this is a meaningful finding for the liver cancer field: it points to hepatocytes as a possible cell-of-origin for NRF2-driven cholangiocarcinoma and suggests SWI/SNF as a druggable vulnerability. The study is technically ambitious, challenging in vivo zebrafish, scRNA-seq, and chemical library screen.

这个table能不能用一个展示figure来展示这种距离关系并且能够直观的highlight出来 这个niche

这个怎么计算，有什么生物学含义，缺少指标figure和不同指标下代表性的图像

eLife Assessment

This set of studies provides important knowledge for the role of zona incerta neurons in motivation and food-seeking in mice. The evidence supporting the findings is convincing, with the use of multiple approaches and nice behavioral procedures to provide converging lines of evidence. Additional analyses and data from the current studies would further strengthen the evidence. This work will be of interest to those interested in motivation and its brain substrates.

Reviewer #1 (Public review):

Summary:

This manuscript by Laura Korobkova and Brian Dias describes an interesting study of the role of GABAergic neurons in the zona incerta (ZI) in incentive motivation for reward.

The authors report that DREADD inhibition of ZI neurons reduced the effort breakpoint in a progressive ratio task, which measures the intensity of incentive motivation to obtain food rewards. In other tests, chemogenetic inhibition did not alter food consumption or memory.

Conversely, DREADD excitation of ZI neurons increased incentive motivation in the progressive ratio task, expressed as a higher breakpoint for food rewards.

Korobkova and Dias report that prior stress exposure to a series of stressors (e.g., forced swim & water submersion, restraint, mild footshock) by itself reduced the breakpoint for food reward under vehicle, though it did not impair the ability to learn an instrumental response. However, DREADD excitation of ZI neurons in previously stressed mice increased the breakpoint to normal levels equivalent to the never-stressed group. This important finding indicates the ability of ZI stimulation to rescue the incentive motivational deficit induced by prior stress.

In fiber photometry studies using vGAT-CRE mice to specifically identify GABA neurons, Korobkova and Dias report that ZI GABA neurons are excited by sensory signals, including neutral cues. However, after reward conditioning, ZI GABA neurons increase their activation to the CS+ cue that predicts reward, but not to the CS- cue that doesn't. ZI neurons also respond in an instrumental reward task during both lever press and reward delivery. The authors conclude that ZI neurons respond to sensory stimuli, but specifically code the motivational significance of reward-related stimuli.

In optogenetic studies, the authors find that ZI GABA neuron stimulation during a reward CS+ enhances motivated responding to obtain reward, particularly in females, but not stimulation outside the CS+. This suggests the ZI stimulation in females may specifically enhance the incentive salience of the CS+, namely the cue's ability to trigger an increase in 'wanting' for the reward. However, that effect was not found here in males.

Altogether, this is a fine contribution to the literature, and the authors deserve congratulations on their study and manuscript.

Strengths:

This is a powerful and creative set of studies that clarifies the roles of ZI neurons in sensory processing and especially in incentive motivation for rewards. The use of multiple methods and test situations to triangulate on reward motivation functions gives a well-rounded perspective on ZI function. The discovery of incentive motivation roles for ZI neurons is intriguing and improves understanding of ZI, which traditionally has been a relatively understudied brain structure. The finding that ZI stimulation may rescue stress-induced deficits in motivation is especially notable and may have therapeutic implications.

Weaknesses:

Minor: This version of the manuscript focuses the introduction and discussion specifically on ZI GABA neurons. The ZI may be primarily GABAergic, but also contains other neurons, and DREADD studies may have used the hSyn promoter, which would impact all types of ZI neurons. Other studies here did more specifically target GABA neurons using vGAT Cre mice and specific targeting. The manuscript might be slightly improved by distinguishing in the discussion a bit more clearly which effects implicate GABA neurons specifically, and which effects might include other neurons too, to more clearly parse out the relative roles of GABA vs broader neuronal populations in ZI.

Reviewer #2 (Public review):

Summary:

This paper describes a study that uses a combination of observational and experimental techniques to investigate the hypothesis that the zona incerta is a neural loci where sensory information is integrated to interpret the motivational value of reward-associated cues. They show that manipulation of GABAergic neurons in this region bidirectionally modulates responding during a progressive ratio test, that activating these neurons recovers motivational deficits incurred by chronic stress, and that they fire in response to reward-associated visual or auditory cues. They also showed that activity in these neurons is not necessary for incentive salience of reward-associated cues, because inactivating them did not prevent Pavlovian-instrumental transfer. However, activating them did enhance responding during the presentation of reward-associated cues in females but not in males.

Strengths:

The study has a very systematic and elegant approach to assess how this region responds first to intrinsic motivation and then to motivation-enhancing effects of reward-associated cues.

Weaknesses:

Males and females are used throughout, but sample sizes are generally too small to make a meaningful interpretation of sex differences (which is not the focus of the study, but is worth bearing in mind). In the last experiment, the lack of discrimination between CS+ and CS- conditions across training for males confounds any interpretation of sex-differences in the outcomes.

The ZI is known to be a region where there is notable convergence of neural inputs from a diverse and heterogenous range of sensory and other cortical inputs. To my knowledge, this is the first study that has directly tested whether it may serve to encode motivational/incentive properties of reward-associated cues. The outcomes are not definitive - it appears that they are sufficient but not necessary. However, this study represents an important first step - the ZI also has notable heterogeneity in the genetic identity of neurons, and properly dissecting the function of ZI microcircuits will likely require characterising function based on more than one molecular marker. This is addressed by the authors in the discussion.

In summary, this study will have a significant impact on our understanding of how motivation is calculated based on complex environmental signals.

Reviewer #3 (Public review):

Summary:

The authors investigated the role of the zona incerta in motivation and cue-reward associations. Using chemogenetic and optogenetic manipulations of the ZI, they altered motivation in cued and uncued variants of the progressive ratio task and rescued deficits in motivation induced by chronic stress. They further use fiber photometry to demonstrate that the ZI tracks the formation of cue-reward associations.

Strengths:

(1) The authors fill an important gap in the literature linking sensory input to motivation via the zona incerta.

(2) The authors demonstrate that ZI tracks cue value rather than just tracking sensory input.

(3) The authors demonstrate that the ZI excitation rescues stress-induced suppression of motivation.

(4) The authors perform several important control tasks, demonstrating that their findings are not a result of alterations in locomotor activity, food consumption, or memory.

Weaknesses:

In Figure 1D and E (inhibitory vs excitatory DREADDS), the control groups in the Gi group appear to have more elevated breakpoints than the control groups in the Gq group, although a statistical comparison between the two is not reported. It is not clear if this is because the two groups were given a different reinforcement schedule, this should be made clearer.

In Figure 1E, it is important to note that although the authors found a significant planned comparison between Gq VEH and Gq CNO, the interaction was not significant, nor were comparisons to mice injected with control virus. Thus, activation of ZI GABA neurons appears to be a relatively weak effect.

In Figure 5, the authors see what is likely a significant difference in lever presses during acclimation between the Gi and GFP groups, which they state is an expected difference. However, it is difficult to see why this would be expected. While Gi:CNO manipulation yielded lower breakpoints in Figure 1D, it did not yield lower FR1 responding for food in Fig S3 (although this was FR1 for food dispenser visits rather than lever press). One reason I ask is that the authors highlight the differences in CS+/CS- between groups, but the biggest difference between groups appears to be in acclimation, which may be driving the group x block interaction.

In Figure 6, the authors demonstrate that optogenetic stimulation during cue light increases the breakpoint in females, but not in males. They suggest that this may be because the males did not sufficiently discriminate the cue light before optogenetic manipulation began. If this were the case, then the authors would need to use "cue discrimination" as a factor to determine if it is a better predictor than sex.

The authors' work demonstrates that chemogenetic inhibition of GABAergic ZI cells reduces uncued motivation for reward but enhances cued responses under extinction. The authors state that this is a paradoxical finding that suggests that the ZI operates within a redundant motivation network. However, a critical difference between the two tasks is that one measures motivation for food while the other measures persistent responding under food extinction, which are not the same process. Thus, a simpler explanation is that ZI inhibition reduces motivation and impairs extinction.

Author response:

Public Reviews:

Reviewer #1 (Public review):

Summary:

This manuscript by Laura Korobkova and Brian Dias describes an interesting study of the role of GABAergic neurons in the zona incerta (ZI) in incentive motivation for reward.

The authors report that DREADD inhibition of ZI neurons reduced the effort breakpoint in a progressive ratio task, which measures the intensity of incentive motivation to obtain food rewards. In other tests, chemogenetic inhibition did not alter food consumption or memory.

Conversely, DREADD excitation of ZI neurons increased incentive motivation in the progressive ratio task, expressed as a higher breakpoint for food rewards.

Korobkova and Dias report that prior stress exposure to a series of stressors (e.g., forced swim & water submersion, restraint, mild footshock) by itself reduced the breakpoint for food reward under vehicle, though it did not impair the ability to learn an instrumental response. However, DREADD excitation of ZI neurons in previously stressed mice increased the breakpoint to normal levels equivalent to the never-stressed group. This important finding indicates the ability of ZI stimulation to rescue the incentive motivational deficit induced by prior stress.

In fiber photometry studies using vGAT-CRE mice to specifically identify GABA neurons, Korobkova and Dias report that ZI GABA neurons are excited by sensory signals, including neutral cues. However, after reward conditioning, ZI GABA neurons increase their activation to the CS+ cue that predicts reward, but not to the CS- cue that doesn't. ZI neurons also respond in an instrumental reward task during both lever press and reward delivery. The authors conclude that ZI neurons respond to sensory stimuli, but specifically code the motivational significance of reward-related stimuli.

In optogenetic studies, the authors find that ZI GABA neuron stimulation during a reward CS+ enhances motivated responding to obtain reward, particularly in females, but not stimulation outside the CS+. This suggests the ZI stimulation in females may specifically enhance the incentive salience of the CS+, namely the cue's ability to trigger an increase in 'wanting' for the reward. However, that effect was not found here in males.

Altogether, this is a fine contribution to the literature, and the authors deserve congratulations on their study and manuscript.

Strengths:

This is a powerful and creative set of studies that clarifies the roles of ZI neurons in sensory processing and especially in incentive motivation for rewards. The use of multiple methods and test situations to triangulate on reward motivation functions gives a well-rounded perspective on ZI function. The discovery of incentive motivation roles for ZI neurons is intriguing and improves understanding of ZI, which traditionally has been a relatively understudied brain structure. The finding that ZI stimulation may rescue stress-induced deficits in motivation is especially notable and may have therapeutic implications.

Weaknesses:

Minor: This version of the manuscript focuses the introduction and discussion specifically on ZI GABA neurons. The ZI may be primarily GABAergic, but also contains other neurons, and DREADD studies may have used the hSyn promoter, which would impact all types of ZI neurons. Other studies here did more specifically target GABA neurons using vGAT Cre mice and specific targeting. The manuscript might be slightly improved by distinguishing in the discussion a bit more clearly which effects implicate GABA neurons specifically, and which effects might include other neurons too, to more clearly parse out the relative roles of GABA vs broader neuronal populations in ZI.

The current version of our manuscript notes “Our chemogenetic manipulations targeted all GABAergic ZI neurons, and emerging evidence suggests molecular heterogeneity within this population that may map onto distinct functional roles (Arena et al., 2024; Wilt et al., 2025). Future work to first profile the molecular heterogeneity of GABAergic cells in the ZI and then using intersectional strategies to target defined ZI sub-populations will be essential to providing a more nuanced view of ZI GABAergic influences on motivation.” Our revision will make sure to discuss newer literature demonstrating cellular heterogeneity of ZI and emphasize that this heterogeneity will provide more nuanced contributions of ZI cells beyond the studied GABAergic population on motivation.

Reviewer #2 (Public review):

Summary:

This paper describes a study that uses a combination of observational and experimental techniques to investigate the hypothesis that the zona incerta is a neural loci where sensory information is integrated to interpret the motivational value of reward-associated cues. They show that manipulation of GABAergic neurons in this region bidirectionally modulates responding during a progressive ratio test, that activating these neurons recovers motivational deficits incurred by chronic stress, and that they fire in response to reward-associated visual or auditory cues. They also showed that activity in these neurons is not necessary for incentive salience of reward-associated cues, because inactivating them did not prevent Pavlovian-instrumental transfer. However, activating them did enhance responding during the presentation of reward-associated cues in females but not in males.

Strengths:

The study has a very systematic and elegant approach to assess how this region responds first to intrinsic motivation and then to motivation-enhancing effects of reward-associated cues.

Weaknesses:

Males and females are used throughout, but sample sizes are generally too small to make a meaningful interpretation of sex differences (which is not the focus of the study, but is worth bearing in mind). In the last experiment, the lack of discrimination between CS+ and CS- conditions across training for males confounds any interpretation of sex-differences in the outcomes.

The goal of our study was not to determine sex differences in motivation mediated by the ZI but rather demonstrate a general role of GABAergic cells in the ZI on motivation. As such, while all our experiments used male and female mice and our statistical analyses did not uncover any sex differences in most experiments, our sample sizes of each sex are too small to definitively make any statements about sex differences. As noted already in our Discussion, the sex-specific cue-utilization behavioral strategy in our optogenetic experiment warrants further investigation as a contributing factor to motivation that may or may not be influenced by GABAergic cells in the ZI. It bears mentioning that, to our knowledge, none of the recently published literature on the role of the zona incerta in learning, memory and appetitive behavior that is cited in this manuscript (including our own prior work) has uncovered sex differences in the contributions of the zona incerta to these behaviors.

The ZI is known to be a region where there is notable convergence of neural inputs from a diverse and heterogenous range of sensory and other cortical inputs. To my knowledge, this is the first study that has directly tested whether it may serve to encode motivational/incentive properties of reward-associated cues. The outcomes are not definitive - it appears that they are sufficient but not necessary. However, this study represents an important first step - the ZI also has notable heterogeneity in the genetic identity of neurons, and properly dissecting the function of ZI microcircuits will likely require characterising function based on more than one molecular marker. This is addressed by the authors in the discussion.

In summary, this study will have a significant impact on our understanding of how motivation is calculated based on complex environmental signals.

Reviewer #3 (Public review):

Summary:

The authors investigated the role of the zona incerta in motivation and cue-reward associations. Using chemogenetic and optogenetic manipulations of the ZI, they altered motivation in cued and uncued variants of the progressive ratio task and rescued deficits in motivation induced by chronic stress. They further use fiber photometry to demonstrate that the ZI tracks the formation of cue-reward associations.

Strengths:

(1) The authors fill an important gap in the literature linking sensory input to motivation via the zona incerta.

(2) The authors demonstrate that ZI tracks cue value rather than just tracking sensory input.

(3) The authors demonstrate that the ZI excitation rescues stress-induced suppression of motivation.

(4) The authors perform several important control tasks, demonstrating that their findings are not a result of alterations in locomotor activity, food consumption, or memory.

Weaknesses:

In Figure 1D and E (inhibitory vs excitatory DREADDS), the control groups in the Gi group appear to have more elevated breakpoints than the control groups in the Gq group, although a statistical comparison between the two is not reported. It is not clear if this is because the two groups were given a different reinforcement schedule, this should be made clearer.

In our revision, we will be sure to insert language re-emphasizing that the Gi and Gq experiments were performed using different reinforcement schedules, FR3 and FR1, respectively.

In Figure 1E, it is important to note that although the authors found a significant planned comparison between Gq VEH and Gq CNO, the interaction was not significant, nor were comparisons to mice injected with control virus. Thus, activation of ZI GABA neurons appears to be a relatively weak effect.

In our revision, we will insert language to acknowledge that reducing motivation after inhibiting GABAergic ZI cell activity is stronger than increasing motivation seen after stimulating the activity of these cells.

In Figure 5, the authors see what is likely a significant difference in lever presses during acclimation between the Gi and GFP groups, which they state is an expected difference. However, it is difficult to see why this would be expected. While Gi:CNO manipulation yielded lower breakpoints in Figure 1D, it did not yield lower FR1 responding for food in Fig S3 (although this was FR1 for food dispenser visits rather than lever press). One reason I ask is that the authors highlight the differences in CS+/CS- between groups, but the biggest difference between groups appears to be in acclimation, which may be driving the group x block interaction.

In a revision, we will revise the language to state that the acclimation difference that is more pronounced in the GFP control group vs the Gi group is to be expected because we had already shown that inhibition of GABAergic cells in the ZI would reduce lever pressing. We will also report a planned comparison of CS+ versus CS− responding that omits the acclimation block, which shows discrimination between sound (CS+) and light (CS−) in the Gi group but not in the GFP group, confirming that the cue effect is not driven by the acclimation difference. We will also note that acclimation responding is non-reinforced and effortful, whereas FR1 dispenser visits (Fig. S3) are reinforced and low-effort, which is why the two measures dissociate.

In Figure 6, the authors demonstrate that optogenetic stimulation during cue light increases the breakpoint in females, but not in males. They suggest that this may be because the males did not sufficiently discriminate the cue light before optogenetic manipulation began. If this were the case, then the authors would need to use "cue discrimination" as a factor to determine if it is a better predictor than sex.

In a revision, we will add an analysis that includes cue discrimination as a factor, to test whether it is a better predictor of the optogenetic effect on breakpoint than sex.

The authors' work demonstrates that chemogenetic inhibition of GABAergic ZI cells reduces uncued motivation for reward but enhances cued responses under extinction. The authors state that this is a paradoxical finding that suggests that the ZI operates within a redundant motivation network. However, a critical difference between the two tasks is that one measures motivation for food while the other measures persistent responding under food extinction, which are not the same process. Thus, a simpler explanation is that ZI inhibition reduces motivation and impairs extinction.

Our revision will include discussion of this important point and tie it into our previous work (Venkataraman et al. 2019 and 2021 – cited in this version) that included extinction-like protocols, albeit in classical (not operant) conditioning protocols.

check this

Are there methods to reduce cost?

Research takes place somewhere in North America, autistic adults are the focus

Idriss argues that when autistic people spend time with each other on a regular basis, they are given an opportunity to build social skills and interpersonal relationships, both within their group and the communities they live in.

name Idriss gave to the group they studied

The community is fluid & seems to have a vary in which individuals attend their gatherings

anthropology term; the researcher joins activities with the group they're observing

Author blended together participant observation, interviews, and making genuine connections as their research methods.

possibly some quantitative research?

Researcher is following ethical guidelines, only doing what the group agrees to

Acomm is a group that is welcoming of all people on the Autism/ ADHD/ ADD spectrum

Acomm is a diverse community

This is a lot like an organization that I'm currently the president of: a Family Readiness Group for military spouses. People are invited to come to events or outings as they are, no matter their age or sailor's rank. It's a community based around supporting each other when our active duty military spouses are away on missions. no obligations or dues, just making connections and uplifting each other.

Here, the author discovered the group's spontaneity allowed for members to open up and be more sociable, something that might be more difficult if they were with a group of neuro-typical people.

the irony here is uncanny

I've known several people with autism, but this line in particular reminded me of a guy I worked with back in 2019. He told me that facial expressions were really difficult for him to read, and it sometimes caused confusion for him and other co-workers.

Haven't seen this word before.. context clues make me think it means something like being on the same wavelength as someone else or seeing eye-to-eye with them.

It seems like when Acomm gathers, the members are able to relax and let their guards down in each other's presence. They don't have to "put on a show" and act "normal" like they would with a group of neurotypical people.

start of research

end of research?

This reminds me of my group of friends - we're all military spouses who have formed a unique community that can lean on each other when we're stressed or need help.

I grew up in a tiny town, and I can say that living in a more populated area is much better for someone who doesn't want to be watched or recognized everywhere they go.

citation!!!

I find it sad that support is cut off for autistic individuals once they become adults, but it makes me happy that organizations like Acomm exist for people to have a safe space.

I have had similar instances where I felt the need to explain that I wasn't intending to rush or push someone to do something.

I thought the concept of studying how groups of autistic individuals act and interact with each other was really interesting.

citation!!!

I thought the concept of studying how groups of autistic individuals act and interact with each other was really interesting.

I find it sad that support is cut off for autistic individuals once they become adults, but it makes me happy that organizations like Acomm exist for people to have a safe space.

I grew up in a tiny town, and I can say that living in a more populated area is much better for someone who doesn't want to be watched or recognized everywhere they go.

I have had similar instances where I felt the need to explain that I wasn't intending to rush or push someone to do something.

This reminds me of my group of friends - we're all military spouses who have formed a unique community that can lean on each other when we're stressed or need help.

end of research?

start of research

It seems like when Acomm gathers, the members are able to relax and let their guards down in each other's presence. They don't have to "put on a show" and act "normal" like they would with a group of neurotypical people.

Haven't seen this word before.. context clues make me think it means something like being on the same wavelength as someone else or seeing eye-to-eye with them.

I've known several people with autism, but this line in particular reminded me of a guy I worked with back in 2019. He told me that facial expressions were really difficult for him to read, and it sometimes caused confusion for him and other co-workers.

the irony here is uncanny

This is a lot like an organization that I'm currently the president of: a Family Readiness Group for military spouses. People are invited to come to events or outings as they are, no matter their age or sailor's rank. It's a community based around supporting each other when our active duty military spouses are away on missions. no obligations or dues, just making connections and uplifting each other.

Here, the author discovered the group's spontaneity allowed for members to open up and be more sociable, something that might be more difficult if they were with a group of neuro-typical people.

Acomm is a diverse community

Acomm is a group that is welcoming of all people on the Autism/ ADHD/ ADD spectrum

Researcher is following ethical guidelines, only doing what the group agrees to

possibly some quantitative research?

Author blended together participant observation, interviews, and making genuine connections as their research methods.

anthropology term; the researcher joins activities with the group they're observing

The community is fluid & seems to have a vary in which individuals attend their gatherings

name Idriss gave to the group they studied

Idriss argues that when autistic people spend time with each other on a regular basis, they are given an opportunity to build social skills and interpersonal relationships, both within their group and the communities they live in.

Research takes place somewhere in North America, autistic adults are the focus

Case#: Patient 8541, male, 10yo at presentation, 9yo at onset

DiseaseAssertion: Stargardt

FamilyInfo: n/a

CasePresentingHPOs:

CaseHPOFreeText: The inclusion criteria for patients were as follows: (1) clinical phenotypes were consistent with retinal dystrophy, and two or more ABCA4 gene mutations were identified by genetic analysis; or (2) clinical phenotypes were consistent with Stargardt disease, and one ABCA4 gene mutation was identified. The diagnosis of STGD was based on visual acuity loss with an atrophic maculopathy with or without yellowish-white flecks. BCVA= 0.92 OU. FAF type 1 ( a localized low FAF signal at fovea surrounded by a homogeneous background with or without perifoveal foci of high or low signal)

CaseNotHPOs:

CaseNotHPOFreeText: Yellowish-white flecks

GenotypingMethod: Analysis by targeted panel sequencing of 256 known retinal disease genes or by clinical exome sequencing of 1651 inherited eye disease-related genes

PreviouslyPublished: n/a

Variant: c.2587_2587+6delGG TAAGC; c.3468C>G p.(Tyr1156*)

ClinVar: n/a

CAID: CA341290648

SupplementalData: supplemental table S2

PMID: 24317291

Gene: ABCA4

Disease: Stargardt

PMID: PMC6548695

Gene: ABCA4

HGNC ID: 34

Case#: Braun Family 5 Proband (from left to right, top to bottom of available pedigrees), male

DiseaseAssertion: Stargardt

FamilyInfo: Proband has 1 affected sister with the same genotype and 1 unaffected, heterozygous brother. Genotypes not provided for parents.

CasePresentingHPOs:

CaseHPOFreeText: "five or more of the following features of ABCA4-associated retinal disease: decreased visual acuity before age 20, decreased visual acuity as the first visual symptom, symmetrical fundus findings, pisciform flecks, beaten metal macular atrophy, bulls-eye maculopathy, peripapillary sparing, vermillion fundus, masked choroid on fluorescein angiography, nummular pigment overlying extensive macular atrophy, central outer retinal atrophy on optical coherence tomography and central scotomas on Goldmann perimetry."

CaseNotHPOs:

CaseNotHPOFreeText:

GenotypingMethod: one plausible disease-causing mutation detected in ABCA4 after assessing the entire coding sequence and canonical retinal splice junctions with automated bidirectional Sanger sequencing using an ABI 3730 sequencer

PreviouslyPublished: n/a

Variant: c.5196+1137G>A; c.4363T>C (p.C1455R)

ClinVar: 438100

CAID: CA26843511

SupplementalData: pedigree in fig s2

another case with 541 variant potentially not in cis with 1038

This paper was listed under this variant in LOVD, but I cannot find it in the main text or supplemental documents.

PMID: 16103129

Gene: ABCA4

Disease: severe retinal dystrophies

Case#: Family AR197 Proband 05, male, 7yo at onset

DiseaseAssertion: arRP

FamilyInfo: parents are het carriers, sibling (06) is affected and has the same genotype

CasePresentingHPOs:

CaseHPOFreeText: "diagnosed by clinical criteria (44) consistent with international standards and confirmed by review of ophthalmic records and retinal photographs. The clinical criteria included visual impairment at early age, progressive loss of peripheral visual functions and typical retinal changes of vascular attenuation, disc pallor and bone spicule accumulation." CF 3 feet OD; CF 2 feet OS

CaseNotHPOs:

CaseNotHPOFreeText:

GenotypingMethod: microarray chip, ABCR-400, PCR

PreviouslyPublished: n/a

Variant: [L541P; A1038V] and[L541P; A1038V]

ClinVar: 99067

CAID: CA226911

SupplementalData: n/a

Case#: Patient 9, female, Mexican, symptoms onset 6 yrs. ago, Mexico City

DiseaseAssertion: IRD

FamilyInfo: parents are non-sanguineous and asymptomatic, they also denied any history related to ocular diseases. Information disclosed that the mother had one stillbirth and three miscarriages, but denied any related diseases/health issues to this child.

CasePresentingHPOs: HP:00305, HP:00080, HP:0000493, HP:0025586, HP:0030329, HP:0012713

CaseHPOFreeText: Proband presented with light sensitivity as well as adaptation difficulties when going from dark-to-light. Right eye was 20/200 and left eye was 20/160 from the visual acuity test. Macular bull's eye appearance. Subnormal rod and cone responses. Peripapillary sparing retina.

CaseNotHPOs: HP:0007737, HP:0000750, HP:0000510

CaseNotHPOFreeText: No afferent pupillary defect. No anomalies in anterior segment.

Genotyping Method: QIAamp DNA Blood Kit was used to extract gDNA and quantification/purity of the sample was found using a NanoDrop 2000 spectrophotometer. 293 genes were sequenced. gDNA was sequenced via Illumina technology. Following, certain sequences were additionally analyzed against a reference genome in order to identify changes and interpret.

PreviouslyPublished: n/a

Variant: NM_000350.3(ABCA4):c.4926C>G (p.Ser1642Arg), NM_000350.3(ABCA4):c.5044_5058del (p.Val1682_Val1686del)

ClinVar: 99332, 99340

CAID: n/a

SupplementalData: Phenotype data in results section as well as figures 1, 2, and 3 showing phenotypic testing results.

This individual appears to be the same as in PMID: 23755871. Same author in both papers

This variant is listed for Stargardt DNAID#070949 in trans with c.3682G>A p.(Glu1228Lys). No phenotype information provided.

PMID: 33214501

Gene: ABCA4

Disease: ABCA4-Related Retinopathy

PMID: 19074458

Gene: ABCA4

Disease: cone-rod dystrophy

Case#: a 46-year-old male

DiseaseAssertion: Stargardt Disease

FamilyInfo:visual acuity loss by his brother and father

CasePresentingHPOs: HP:0007663

CaseHPOFreeText: visual acuity measured 20/400 bilaterally

CaseNotHPOs: NR

CaseNotHPOFreeText:NR

Genotyping Method: ABCA4 microarray (ABCR5000 chip)

PreviouslyPublished: NR

Variant: c.5714+5G>A

ClinVar: 99403

CAID: CA227338

SupplementalData:NR

Case#: Patient 66yo, M, Japan, Symptoms in teens

DiseaseAssertion: Stargardt (STGD) Heterozygous, Autosomal recessive

FamilyInfo: No family with similar symptoms

CasePresentingHPOs: HP:0008002, HP:0007722, HP:0000610, HP:0000602

CaseHPOFreeText: Ring shaped paracentral scotoma

CaseNotHPOs: HP: 0030329, HP:0000608, HP:0000493

CaseNotHPOFreeText: Well preserved foveal function, well preserved retinal, and normal thickness RPE

Genotyping Method: N/A

PreviouslyPublished: 4 other polymorphisms

Variant: c.839T>C p.Met280Thr Heterozygous

ClinVar: 1349490

SupplementalData: see tables 1 for gene variant and found polymorphisms

Case#: single case, 37-year-old male, ethnicity not specified although family originally from the Middle East, examined in the UK

DiseaseAssertion: STGD

FamilyInfo: No history of consanguinity. No history of inherited retinal disease, poor vision or colour vision disturbance. Father had recent diagnosis of chronic central serous retinopathy, not consistent with STGD

CasePresentingHPOs: n/a

CaseHPOFreeText: Late-onset Stargardt disease with slowly progressive phenotype. The patient present with a 3-year history of decreased vision in the right eye that had recently significantly worsened. Visual acuity was 6.24 in right eye and 6/6 in left eye. Fundus examination reveled scattered atrophy and pisiform fundal flecks in both eye, right worse than left. Fluorescein angiography showed a silent choroid and partial bull's eye maculopathy, right worse than left. OCT showed loss of photoreceptors in both eyes and partial central sparing in the left eye. Photopic and scotopic ERG showed reduced amplitude of responses in the right eye and lower range amplitudes in the left eye, normal implicit times in both eyes. Pattern ERG and multifocal ERD showed central retinal dysfunction with preserved peripheral function. No change in vision or retinal appearance over the next 14 months of follow up.

CaseNotHPOs: n/a

CaseNotHPOFreeText: No night vision symptoms. No history of retinotoxic drug exposure.

Genotyping Method: Next-generation sequence analysis with the Oxford Genetics Testing Laboratory Macular Gene Panel.

PreviouslyPublished: Thr829Met missense mutation had been previously reported in an individual with autosomal recessive retinitis pigmentosa, but not previously associated with STGD phenotype.

Variant: ABCA4 NM_000350 c.5882G>A, p.(Gly1961Glu) c.2486C>T, p.(Thr829Met)

ClinVar: n/a

CAID: CA958261, CA119132

SupplementalData: n/a

PMID:39215425

Gene: ABCA4

HGNC ID: 34

Case#:1 this ia family but the 20 year old son is the firs tone spoken about a herdirtary eye disease, shows phenotype for years till syptmos worsned with age

DiseaseAssertion: Table 1 The summary of the clinical assessment of IRD patients’ family in this research fro there down they did a whole pannel on the family

Pedigree one can be fore form the beggginnings of case presention section?

CasePresentingHPOs: suffered from tunnel vision and blurry night vision began 13 years ago

CaseHPOFreeText:NA

CaseNotHPOs:NA

CaseNotHPOFreeText:NA

Genotyping Method:NA

PreviouslyPublished:NA

Variant:NA

ClinVar:

CAID:NA

SupplementalData:NA

Inheritance pattern Autosomal Recessive

Per ClinVar entry, this variant was associated with this paper. however, after reading through the genotypes, this variant was not found. Likely this paper was mentioned to support the statement that "Loss-of-function variants in ABCA4 are known to be pathogenic"

PMID: 27208204

Gene: ABCA4

Disease: IRD

This variant is found in table S2A to have an OR of 47.7 with a CI (4.71-2311.43).

Perspective of the usefulness of AI and how it can be beneficial

AI used for photo editing or altering images

This supports that AI is not always accurate

Creates quicker ways of doing things and planning.

Perspective to use: advances in AI "could spell the end of the human race."

Knowledge is meaningful information

This supports that AI can't replace emotions, feelings and common sense. Won't replace our ways of learning

The Post-Roe Surprise: More Abortions<br /> by [[Reveal]] in Mother Jones<br /> accessed on 2026-08-02T21:36:29

ROBERT CHARLES;

gunmemorial; 4ta victima; John F. Lally

AI used in picture editing as an example.

Interesting metaphor and perspective

Supports idea that AI is not always accurate.

This takes out emotion in regards to how humans problem solve.

Will AI replace human thinking and education?

eLife Assessment

This valuable paper reports on a measure of flexible brain state engagement, derived from fMRI, as a predictor of cognitive control. One strength of the study is the use of external datasets for validation and replication. The results are solid, although some may benefit from further explanation.

Reviewer #1 (Public review):

Summary:

This paper uses three different datasets to study the relationship between the standard deviation of dynamic brain state time series (state engagement variability or SEV) and measures of cognition. Results show associations between SEV and cognitive measures, with stronger associations in patients than controls (at least for inhibition).

Strengths:

Strengths include the use of innovative dynamic approaches to study cognition and the validations across three independent datasets.

Weaknesses:

With a highly innovative approach, it can be challenging to provide enough context for the reader to understand and interpret the results. In particular, the paper would benefit from:

(1) More detail on the brain state calculation, multiple comparison control, and added benchmarking of the novel summary SEV measure.

(2) Guidance on the interpretation of relatively low prediction performance, negative t-statistics, and more broadly regarding the justification for the multi-step approach going from 4 brain states to 1 SEV to a network of edges.

(3) Removal of the moment-to-moment alignment results given the circularity of the edge time series extraction with overlapping contributions to SEV and cognitive control time series.

(4) Adjustment of text to avoid causal interpretations and to reduce the emphasis on transdiagnostics.

Major Points:

While the brain states were developed in prior work, SEV is a new metric and therefore warrants careful benchmarking in terms of test-retest reliability, sensitivity to scan length/quality, and associations with demographic variables like age and sex (which do not appear to be controlled for in analyses).

Although the external validation approach is appreciated, the prediction performance is pretty low (predicted-observed correlation 0.17-0.3). It would be good to also report other metrics of performance, such as balanced accuracy.

The steps in the paper are somewhat convoluted by going from 4 brain states to 1 SEV, back to specific FC networks. This makes the paper a bit complex and difficult to interpret. It would be helpful to provide a clear justification for these steps and/or a figure to orient the readers.

Many results are reported in the manuscript, and it is unclear whether/what multiple comparisons control was adopted where.

The moment-to-moment change section tries to test whether inter-individual variation in SEV maps onto cognitive control, which is very interesting. However, both measures were operationalized using edge-timeseries calculated from the same data with shared inputs (as shown in Figure 4B). As such, the 'alignment' (i.e., correlation) between resulting time series appears somewhat circular given that it is likely driven by the shared inputs. More broadly, edge timeseries were summed across edges (and subtracted between edges with positive and negative CPM associations), which further complicates their interpretability in the context of 'cognitive control'. I would recommend removing this section or using behavioral data to quantify cognitive control.

The descriptions of how brain states were derived are unclear. In line 466, what do 'these fMRI data' refer to? Was the least-squares regression performed across subjects (given that it results in one beta value per time point)? Was this performed as a multiple regression and - if so - what was the collinearity between brain state inputs?

Reviewer #2 (Public review):

Summary:

A relatively new measure of flexible brain state engagement (SEV - State Engagement Variability) is used here. It simply measures time-to-time variation in brain activity in terms of how it matches pre-specified motifs of activity. This metric seems to be predictive of behavioural data measuring cognitive control abilities. This was found to be the case in two independent datasets with different (though related) behavioural measures.

Strengths:

Use of multiple datasets is a clear strength. The use of both replication and out-of-sample model prediction is another.

Weaknesses:

(1) It is not clear to me how specific the SEV metric is for telling us about brain state engagement flexibility. Resting state fluctuations have been described as quasi-periodic changes that can be mapped onto "states", but the fluctuations could easily be a reflection of vascular flow, which may indirectly correlate with cognition.

(2) If SEV is calculated using other state descriptors (e.g. a random parcellation of the brain into 4 networks) - would the result still hold? Or are the motifs important (this would rule out, to some extent, the vascular argument from (1) above)?

(3) Figure 1 confused me a little. Why not show all the combinations (patient v full sample), inhibition vs shift, and main vs validation? Instead, a subset of 4 was selected?

(4) The inhibition/patient/main correlation seems to be driven by 4 patients with particularly high inhibition measures?

(5) Why is SEV negative in some cases (e.g., Figure 1) if it's a std measure? Has it been demeaned or orthogonalised wrt another variable?

(6) The external analysis is great, but why should the model predict a relationship between SEV and inhibition if the claim is that it is only true for patients? Why would it only be true for patients in the first place?

(7) I can't get my head around the results shown in Figure 3. How can one have both positive and negative correlations being significant or meaningful in the same pairs of networks? I think this set of results could benefit from more explanation.

(8) I struggled with Figure 4 analysis. What is the SEV network? How do we know that it is specific enough to the SEV concept? Looking at co-fluctuations with the cognitive network, are we not simply looking at the old anti-correlation between the default mode and the rest of the brain (I note that the correlations in the y-axes of Figure 4 are negative)?

eLife Assessment

This valuable study aimed to explore whether sensory representations in the cortex reorganize on the same timescale over which behavioral changes first emerge. Convincing evidence is presented to show that reward-dependent plasticity takes place in the mouse barrel cortex within a single recording session as mice learn a whisker detection task. However, the evidence supporting the authors' proposal that this rapid reorganization may involve spontaneous reactivation of neurons that gain stimulus responsiveness during training is incomplete. The work will be of interest to systems neuroscientists studying cortical circuits and learning.

Reviewer #1 (Public review):

Summary:

The paper submitted by Renard et al. seeks to capture the moment when learning occurs and to identify the associated changes in neuronal activity within cortical circuits. Specifically, the study aims to test whether sensory representations in the cortex reorganize on the same timescale over which behavioral changes first emerge.

To address this question, the authors developed a new behavioral paradigm in which mice were first trained on an auditory detection task and then introduced to whisker stimulation, which they learned to associate with reward. This design allowed mice to form a new whisker-reward association within a single behavioral session, enabling the authors to track learning-associated neuronal changes during the course of the experiment.

Using pharmacological and optogenetic interventions, the authors first show that learning depends on the whisker somatosensory cortex. They then combined the task with longitudinal two-photon calcium imaging to examine real-time changes in neuronal representations that accompany improvements in task performance over trials within a session and across days. By applying a range of analytical approaches, they show that learning induces a rapid reorganization of sensory cortical representations over tens of trials, on the timescale of minutes. They further propose that spontaneous reactivation of neurons during the task may contribute to these representational changes during learning.

Strengths:

(1) Overall, the experiments are thoughtfully designed, well controlled, and clearly presented. The conclusions are generally well supported by the data. The manuscript is clearly written, and the Discussion acknowledges potential caveats while outlining future directions.

(2) A major strength of the study is the design of a new learning paradigm in which head-fixed mice rapidly form a new sensory-motor association within a single session, on the timescale of minutes. This offers a unique opportunity to track real-time changes in neuronal dynamics associated with learning during a single recording experiment.

(3) Taking advantage of this behavioral design, the authors show that learning induces rapid reorganization of sensory cortical representations. They also report an increase in spontaneous reactivation of neurons that gained stimulus responsiveness during training, and propose that these reactivations may contribute to rapid representational reorganization. These findings provide important insights into the neural dynamics associated with learning.

Weaknesses:

(1) The authors propose that spontaneous reactivation mediates rapid reorganization of neuronal representations and thereby supports rapid task learning. However, as they also acknowledge in the Discussion, the present study does not directly test a causal role for these reactivations in facilitating representational changes or behavioral improvement.

(3) The authors show reorganization of neuronal representations even on the first day of training with the new whisker task. However, because there is no explicit control for natural representational drift, it remains unclear to what extent these changes reflect learning-related reorganization rather than spontaneous day-to-day drifts in neuronal responses.

Reviewer #2 (Public review):

Summary:

Renard, Foustoukos and colleagues present a study of rapid sensorimotor learning in the mouse barrel cortex. Head-fixed water-restricted mice already trained on an auditory detection task are introduced to a novel C2 whisker stimulus, and the authors show that reward-paired mice acquire the whisker-lick association within a single behavioral session, with the two groups (rewarded vs non-rewarded) diverging behaviorally within ~22 whisker trials and ~14 minutes. Both pharmacological inactivation of wS1 across Days 0/+1/+2 and optogenetic inactivation on Day 0 impair whisker-guided performance, while fpS1 manipulations do not, establishing that wS1 activity is required for whisker-guided behavior during the initial learning period. Longitudinal two-photon imaging of GCaMP6f-expressing L2/3 neurons across five days (-2 to +2 relative to whisker introduction) reveals a bidirectional, reward-dependent reorganization of population responses to passive whisker stimuli: rewarded mice show enhancement, non-rewarded mice show suppression. The authors use a logistic-regression decoder trained to discriminate pre- vs post-learning passive trials and then project Day 0 active whisker trials onto this learning axis; the projection rises monotonically across Day 0 in R+ mice and is significantly correlated with behavioral performance, with no such trajectory in R- mice. Finally, the authors detect reactivation events during catch trials by template-matching to the average passive whisker response, and show that on Day 0, the neurons most positively modulated by learning (LMI-positive) participate in these reactivations more than LMI-negative neurons in R+ but not R- mice. The authors interpret this as evidence that online, reward-gated reactivations may act as an upstream selection mechanism for which neurons undergo learning-related plasticity, operating on the minutes-timescale of within-session learning. There is much to like in this paper, with some moderate-to-major concerns that could largely be addressed with re-analysis or re-framing.

Strengths:

The single-session learning paradigm is a key aspect of this paper, given the rapid learning observed. Coupled with the R+ and R- design, there's a lot to like with the behavioral approach. The bidirectional response change across these R+ and R- groups (enhancement vs suppression) is also a nice finding.

The causal manipulations demonstrate that the imaged region is used during the task. By doing both pharmacological and optogenetic inactivation, each with a control in the spatially adjacent region (fpS1), the authors make a strong case that wS1 activity is necessary for whisker-guided behavior during the initial learning period (though see below about the limitations of the current approach).

The longitudinal two-photon imaging of the same L2/3 neurons across five days underlies essentially every neural analysis in the paper and enables the single-cell LMI and population-trajectory analyses.

The pathway-specific analysis in Figure 3 - figure supplement 2 is very interesting, but not much time is spent on it (lines 151-155). The dissociation between wS2-projecting neurons (which show learning-related enhancement in R+ and suppression in R-) and wM1-projecting neurons (which do not) is (in my opinion) a nice instance of projection specificity - it also aligns with the known routing of task-relevant whisker information through the wS1→wS2 pathway. I would encourage the authors to motivate this experiment in the main text rather than leaving it all to the discussion (lines 256-262).

The methods are generally well documented and easy to follow.

Weaknesses:

(1) Conflation of de novo association learning with generalization from auditory pre-training.

All mice have already learned a task structure with the auditory task - "detect the salient sensory cue → lick → reward". Under these conditions, the rapid emergence of licking to the whisker stimulus could reflect either de novo formation of a whisker-specific association or generalization of an instrumental policy to a novel salient cue. The manuscript frames the result as the former ("acquisition of a novel sensorimotor association"), but the experiment cannot distinguish between the two alternatives. This distinction between de novo learning and generalization may have a meaningful impact on the interpretation, though it doesn't impact the specific results. It would be helpful for the authors to discuss the two possibilities and generally consider the contribution of generalization from auditory pre-training to Day 0 performance.

Relatedly, the R- group is introduced (lines 69-74) and later used (lines 244-247) as a passive-exposure control that rules out representational drift. While R- group is an important control for repeated whisker stimulation and task context, it does not appear to be a pure passive-exposure control: Figure 1B shows that on Day 0 the mice lick more to the R- stimulus than with no stimulus and then extinguish that licking by Day 1. Thus, one possibility is that R- mice actively learn to suppress licking to an unrewarded stimulus (whisker) in a context where other stimuli (auditory) remain rewarded. This would be a different cognitive operation (response suppression) from a purely passive exposure condition. The manuscript therefore lacks a true passive-exposure baseline, and several claims that rely on R- as such a baseline (including that bidirectional changes are reward-driven rather than reflecting passive drift, lines 244-247) need to be reframed.

(2) The inactivation experiments establish that wS1 is necessary on Day 0, but they cannot separate detection, acquisition, and expression.

Both the muscimol manipulation (whole session, Days 0/+1/+2) and the optogenetic manipulation (0.1 s before stimulus onset through the 1 s reporting window) silence wS1 during the moments when the whisker stimulus must be detected for a successful trial. Under these conditions, impaired performance could reflect that the animal cannot detect the stimulus, cannot express the learned response on that trial, or cannot acquire the association. These are causally distinct processes, and the manuscript currently treats them as equivalent.

Specifically, on Day +1 of the opto experiment (light off), do mice learn at the same rate as a naive Day 0 cohort (e.g., the R+ imaging mice on Day 0), or is performance already higher than the naive group? If higher than the naïve group, this would suggest that there is learning occurring and would suggest that something that may have been acquired during Day 0 inactivation, even if it could not be expressed.

(3) The interpretation of the LMI-participation correlation is complicated by the peaked LMI distribution and neuron-level pooling.

Two related issues arise from the results shown in Figure 4I. First, the LMI distribution in Figure 3F (and visible in 4I) is sharply peaked near zero. The reported r = 0.24 in R+ mice is therefore difficult to interpret biologically because the distribution is dominated by near-zero-LMI neurons and the slope may be disproportionately influenced by neurons in the tails. The key claim is better tested by comparing significantly LMI-positive, LMI-negative, and non-modulated neurons. The authors do address this in Figure 4J - showing that participation rate rises across days for significantly LMI-positive R+ neurons (p = 5×10⁻⁴) but not for LMI-negative neurons (p = 0.05) - but this analysis is not the lead result. To my understanding, Figure 4J is more interpretable and should be the key piece of data supporting their claim.

Second, the p-value of p = 1×10^-41 in Figure 4I comes from treating thousands of neurons pooled across 19 mice as independent observations. Neurons within an animal are correlated through shared behavioral state, shared imaging session, and circuit-level interactions, so it would be helpful to consider a different statistical unit of comparison (FOV, animal, etc). For example, a linear mixed-effects model with mouse as a random effect could work.

(4) The reactivation-LMI relationship is partially circular, and the framing in the abstract could be more constrained.

The "reactivation template" is the trial-averaged passive whisker-evoked population vector from each session, and reactivations are detected as moments in catch-trial activity that correlate with this template above a shuffled threshold. This approach is reasonable, but it means that the reactivation-LMI relationship is not fully independent of template construction, and the framing in the abstract blurs that line. LMI-positive neurons are defined as neurons whose passive whisker-evoked responses increase from pre- to post-learning. Therefore, neurons with strong whisker responses, or neurons that become stronger components of the whisker-evoked template across learning, may be more likely to contribute to template-matching events by construction. Thus, the LMI-participation relationship could partly reflect template weighting or sensory-response amplitude, rather than showing that reactivation events selectively recruit neurons for future learning-related plasticity. It would be helpful and more reassuring if the authors could control for each neuron's whisker-template weight, baseline whisker responsiveness, and overall calcium event rate when relating LMI to reactivation participation.

A complementary unsupervised approach could also help: rather than starting from the whisker template, one can derive co-activity assemblies directly from spontaneous activity (e.g., via PCA or ICA on the catch-trial population activity), and then ask, separately, whether any of these assemblies overlap with the whisker ensemble. The interesting test is then whether whisker-like assemblies become more frequently expressed across Day 0 in R+ but not R- mice, and whether LMI-positive neurons are preferentially loaded onto these whisker-like assemblies. This logic inverts the current pipeline and can be complementary to the current analysis. By identifying structure in nominally spontaneous activity first and then comparing to the whisker response, this could help avoid the circularity in which the template both defines the events and contains the cells being tested. The Figure 4 - figure supplement 1B partial-correlation analysis is a step in this direction but addresses only spontaneous firing rate, not template coupling. Without such a complementary approach, the authors may want to clarify that the reactivation detection is anchored to a template defined in part by the same cells whose participation is being tested.

(5) The reactivation-as-selection-mechanism interpretation is not supported by the current data.

The Discussion (lines 278-281) acknowledges that the authors have not shown necessity, but the end of the intro and part of the discussion (Lines 275-277) frame reactivations as a "reward-gated selection mechanism" for plasticity. An equally plausible alternative is that neurons whose synaptic inputs or intrinsic excitability have been potentiated by reward-driven learning will simply co-fire more often during quiet periods - meaning reactivations would be a consequence of plasticity that has already occurred rather than a mechanism that selects which neurons to potentiate. The current data cannot distinguish these.

A separate concern is the use of the term "spontaneous." The authors' usage is defensible in one sense - catch trials are stimulus-free, so the activity is not externally driven. However, "spontaneous" in the reactivation literature typically connotes offline, internally generated activity during quiet wakefulness or sleep, which carries different implications for plasticity than activity during active task engagement. Catch trials in this paradigm occur within the behavioral session, with the animal still engaged in the task, potentially anticipating reward or licking. The authors should either acknowledge this distinction in the text or qualify the term - "within-session" or "inter-trial" reactivations would be more accurate and would avoid borrowing the conceptual weight of the offline-replay literature.

The authors should also clarify whether catch-trial activity around licks (false alarms, anticipatory licks) is excluded from the reactivation analysis, and whether reactivation rates depend on recent reward, recent whisker trial outcome, or behavioral state. Specificity controls - template-matching with shuffled templates and with auditory templates - would help establish that detected events reflect whisker-specific patterns rather than generic high-coactivity moments.

(6) Motor, lick, and behavioral-state confounds in the neural analyses are not fully addressed.

I have two specific concerns. First, for the Day 0 active-trial projection, mean whisker reaction times in Figure 1 - figure supplement 1G are around 350-500 ms, but the distributions extend into the 0-300 ms analysis window. The correlation between the projection trajectory and the behavioral learning curve (Figure 4E, lines 196-198) is the key piece of evidence that the neural shift tracks learning. However, on hit trials the lick may fall within or close to the analysis window, so a motor confound could in principle contribute to the rising projection. The authors could repeat the projection using an earlier/shorter window, exclude trials with early licks, or regress out lick timing. It would be helpful to better understand whether this effect is, in part, driven by licking activity.

Second, the central evidence for representational reorganization (Figure 3) rests on a post-session passive epoch in which 50 whisker stimulations are delivered after "task disengagement" (lines 131, 387-389). The concern is that the brain state during this epoch is unlikely to be matched across groups or across days. R+ mice receive additional water rewards on whisker trials, whereas R- mice receive rewards only on auditory trials. This could lead to systematic differences in satiety, arousal, and disengagement state during the passive block. Because cortical sensory responses are strongly modulated by arousal, some of the apparent learning-related enhancement (R+) or suppression (R-) of passive whisker responses across days could reflect systematic state differences during the passive epoch rather than plasticity. The disengagement criterion ("stopped licking in all trial types") is also qualitative - no consecutive-miss or time-window threshold is specified - so the epoch may begin at slightly different behavioral states across mice. To resolve this, the authors could (i) specify the disengagement criterion quantitatively and (ii) compare pupil diameter and whisker self-motion (if available) across R+ vs R- and across days during the passive epoch.

Reviewer #3 (Public review):

This is a methodologically sound manuscript and provides reasonably interpretable results. While being appropriate, they do not seem to bring entirely novel concepts; nevertheless, most of my comments concern the calibration of the interpretive claims rather than the quality of the data.

Strengths:

(1) Longitudinal within-subject imaging:<br /> Tracking the same layer 2/3 neurons across learning allows the bidirectional effect (enhancement in R+, suppression in R-) to be measured within identified cells rather than inferred across cohorts.

(2) Appropriate behavioural controls:<br /> The R+/R- design controls for repeated sensory exposure, and maintaining rewarded auditory trials in both groups controls for engagement and arousal, arguing against disengagement as the source of the R- effect.

(3) Convergent causal manipulations:<br /> Muscimol and optogenetic inactivation both abolish acquisition and include an adjacent control region (fpS1); the temporally restricted optogenetic result partially addresses the concern (Hong et al., 2018) that sustained inactivation may destabilise downstream circuits.

(4) Convergent analyses:<br /> Single-cell learning modulation indices, population similarity measures, and a trial-resolved decoder projection onto a naïve-to-expert axis provide consistent evidence that representational change is concurrent with behavioural acquisition.

(5) Projection-specific resolution:<br /> Retrograde labelling shows learning-related changes in wS2-projecting, but not wM1-projecting neurons, consistent with preferential routing of task-relevant signals through the wS1 to wS2 pathway.

(6) Mechanistically motivated reactivation analysis:<br /> Relating rapid, reward-dependent plasticity to spontaneous reactivations on a timescale of minutes is an original use of the single-session paradigm.

Weaknesses and points requiring clarification

(1) The stimulus is not strictly novel: Passive whisker stimulations were delivered on pre-training Days -2 and -1, so what changes on Day 0 is the stimulus-reward contingency rather than the stimulus itself. This resembles contingency reassignment with reversal-like properties (and possible habituation or latent inhibition) rather than de novo learning, and the licking response is already established during auditory training. The framing should be qualified accordingly.

(2) Barrel cortex dependence should be stated more narrowly: The data show that acute wS1 suppression prevents acquisition of this task, not that whisker detection in general requires barrel cortex; cortical dependence varies with task and manipulation (Hong et al., 2018; Ryan et al., 2022 vs Miyashita and Feldman, 2013). The near-threshold explanation would require psychometric or stimulus-intensity data.

(3) The passive block carries confounds: It is acquired after task disengagement, when satiety, arousal, and reward history differ across groups and days. The authors should report within-block response adaptation and the robustness of the main results to early versus late passive trials. Additionally, could passive presentation of the stimulus without reward delivery lead to devaluation of the stimulus, leading to additional behaviour and plasticity changes which are not addressed?

(4) Some statistics appear to be neuron-level rather than animal-level:<br /> Very small p-values (e.g., the LMI-participation correlation r = 0.24, p on the order of 10^-41) suggest thousands of non-independent neurons treated as independent samples, risking pseudoreplication. Central claims should rest on hierarchical or animal-level statistics with effect sizes.

(5) The cosine-similarity decrease in R- animals needs clarification:<br /> Because cosine similarity is scale-invariant, uniform suppression would leave it largely unchanged; the observed decrease therefore implies heterogeneous suppression, reduced signal-to-noise, or increased variability, and the favoured interpretation should be stated.

(6) The decoder requires cautious interpretation:<br /> Training on passive trials and applying to active Day 0 trials could introduce a behavioural-state domain shift. The meaning of positive and negative values in Figure 4C should be defined, and near-zero early projections reflect the classifier boundary rather than a biological baseline.

(7) The reactivation analysis is the least conclusive and is susceptible to circularity:<br /> The template and the LMI are both derived from the passive whisker response, predisposing responsive neurons to register as reactivation participants, and the Day 0 template is obtained after learning.

Leave-one-cell-out and pre-learning templates, cell-specific templates, and tests of whether reactivations predict subsequent trial responses would strengthen the claim; causal disruption would ultimately be required.

(8) Figure, sample-size, and specificity points:<br /> The positive LMI shift in R+ animals is less visible than the R- shift in Figure 3F; the optogenetic cohort is small (n = 6 per group); and confirming that auditory detection was preserved during wS1 inactivation would establish whisker-specificity.

(9) The comparison to prior work is overly broad:<br /> Banerjee et al. (2020) and Chéreau et al. (2020) are reversal learning and discrimination paradigms and may not be equated with simple whisker detection; the defensible novelty claim is the trial-resolved tracking within the first session and its concurrence with online reactivations.

eLife Assessment

This valuable study describes a simple and robust approach for estimating information-limiting noise by splitting neural populations and comparing estimator values. The authors report more accurate and robust results compared to previous methods. The evidence for the robustness of the method is currently incomplete; some concerns regarding bias need to be resolved, and additional tests need to be provided on how the number of trials and neurons affect performance.

Reviewer #1 (Public review):

The authors address a difficult and well-known problem in systems/computational neuroscience: how to estimate the magnitude of "information-limiting" noise. Existing approaches (direct Fisher-information estimation, decoding + Cramer-Rao, and large-N extrapolation) are data-hungry and unstable, which has left the field with conflicting empirical estimates across systems.

The central proposal - "split-trial analysis" - is simple and appealing. The recorded population is randomly partitioned into two non-overlapping halves; a decoder (continuous case) or classifier (binary case) is trained on each half using the same trials; and the covariance of the two halves' decoding errors is used to estimate the variance of the information-limiting noise. There is a clean mathematical derivation to support this conclusion (although there are a couple of mathematical errors in the methods section that should be fixed to avoid confusion on the part of the reader).

They benchmark the method in simulation against three prior methods (Moreno-Bote et al. 2014; Rumyantsev et al. 2020; Kafashan et al. 2021) and report substantially better sample efficiency, lower bias, and greater robustness. They then apply the method to three datasets: (1) mouse head-direction cells (Ajabi et al.), (2) mouse V1 (Stringer et al.), and (3) macaque PFC during a saccade task (Bartolo et al.).

This is a strong and timely contribution. The core idea is elegant, and the method appears to be more practical than existing alternatives in the finite-data regime that real experiments occupy. The three applications are well chosen, and each yields a non-trivial, biologically interpretable result. I am strongly supportive of the potential of this paper.

That said, the paper makes several strong empirical claims - most notably that prior V1 estimates were substantial overestimates, and that PFC information-limiting noise is temporally redundant - and the central estimator rests on an independence assumption whose finite-N validity is only partially characterized. Before these claims can be considered well supported, I would like the authors to address the following:

Major Points:

(1) The method relies on a key independence assumption that may not always be satisfied in the regime of finite neurons and trials. The author's main idea is to decompose the residuals of two decoders as follows:

X1 = delta + phi1<br /> X2 = delta + phi2

The covariance is equal to the scale of information limiting noise, Var[delta], plus three terms:

Cov[X1, X2] = Var[delta] + Cov[delta, phi1] + Cov[delta, phi2] + Cov[phi1, phi2].

We can define phi1 as the part of X1 that is orthogonal to delta and likewise define phi2 as the part of X2 that is orthogonal to delta; thus, the cross terms evaluate to zero, and we are left with:

Cov[X1, X2] = Var[delta] + Cov[phi1, phi2]

Now the authors introduce an assumption that Cov[phi1, phi2] = 0. This leaves us with Cov[X1, X2] = Var[delta], but the question is: when is it justified to assume that Cov[phi1, phi2] = 0? For example, it is possible that

phi1 = c(N) * z + e1<br /> phi2 = c(N) * z + e2

where z is another shared noise dimension that is not information limiting and e1 and e2 are truly independent. Here, c(N) is a constant that goes to zero as the number of neurons used to train the decoder, N, goes to infinity. Thus, in the limit of having very large neural populations at hand for the analysis, the author's assumption of Cov[phi1, phi2] = 0 can be justified. If the authors agree with this analysis, it would be nice to (a) flesh it out and include it in the methods / supplementary notes, and (b) to analyze in simulation how good this approximation is in finite N regimes. I suspect that the assumption works in finite N regimes if noise is low-dimensional, but that if there are many additional dimensions of correlation (i.e. many z's above), you will need a very large number of neurons before Cov[phi1, phi2] approaches zero.

Along these lines, another worthwhile analysis would be to report outcomes when the neural populations are sub-sampled further. Intuitively, it should fail once you subsample to only a handful of neurons, e.g. 3, but I'm curious where the breaking point is and whether the decline is graceful.

(2) In point 1, I raised the question of how the method behaves with a finite number of neurons. Another worry is that there is a finite number of trials. In particular, if you train two decoders on the same trials, I would worry that non-information-limiting fluctuations in those trials would induce correlations in the decoders that then would show up as correlations on the held-out test set. A more conservative approach would be to split trials into three disjoint subsets: a training set for decoder A, a training set for decoder B, and a common test set used to compute Cov[X1, X2].

As a concrete example, suppose that on the particular trials used for training, the animal happened to be more aroused when theta = 1 and less aroused when theta = 0, and that arousal added a fluctuation on top of the neural response. This arousal-related signal is not information-limiting - it would average away given enough trials - but because both decoders are fit to these same trials, each one adjusts its weights to partially discount the same spurious high-arousal/low-arousal trend. Their weights are now distorted in a correlated way, so when both are applied to the shared test set, their errors covary, and the method reads this shared-training artifact as information-limiting noise.

I think this dynamic should be acknowledged in the text and clarified in more detail. Ideally, simulations could be done to estimate how many trials are needed to average out this sort of confound, and similar to the suggestion in point 1 above, I would be interested in seeing what happens when the authors sub-sample trials before running their analysis. Together with point 1, the feedback is that I'd like to see more about "how many neurons and how many trials" are needed in order to trust your results. Similarly, are there diagnostics or resampling methods (e.g. bootstrapping) that could be helpful for a practitioner to know if they have enough neurons/trials?

(3) Unless I've fundamentally misunderstood something, there is an error on page 17 in the methods. There we find sigma2 = Var[delta] = ... = Cov[phi1, phi2], but I believe this is meant to be Cov[X1, X2]. Indeed, the method assumes that Cov[phi1, phi2] = 0, as discussed in point 1.

Additionally, on page 4, the authors introduce the main quantity as Cov[\hat{theta}_1, \hat{theta}_2] instead of Cov[X1, X2]. However, if theta is changing from trial to trial, then these two quantities are not technically equal to each other, so it would be more accurate to write down the conditioning on theta. That is, assuming conditionally unbiased decoders, Cov[X1, X2] = Cov[\hat{theta}_1, \hat{theta}_2 | theta] for a fixed theta.

More generally, I found it hard to wrap my head around the underlying math on my first read through the paper. The polarization identity, 1/4 * (Var(X1 + X2) - Var(X1 - X2)), seems like a very roundabout way to derive the method. This identity is very helpful for the deconvolution extension, but I would have thought that a simpler and more straightforward derivation would have just used the expansion, Cov[X1, X2] = Var[delta] + Cov[delta, phi1] + Cov[delta, phi2] + Cov[phi1, phi2], as I did in point 1. I suggest the authors revise the mathematical presentation for clarity.

Minor Points

(1) A very nice feature of the authors' method is that they make no parametric assumption on the distribution of noise. This is in contrast to Kanitscheider et al. [12]'s finite-sample bias correction using the inverse-Wishart distribution of $\hat\Sigma^{-1}$, which is derived under an assumption of multivariate Gaussianity. I think it is worth adding a sentence to highlight this feature of the model.

(2) Statistical inference claims (across sessions and population sizes) are supported by reported s.d.'s but no formal tests or confidence-interval-based comparisons. Given that several claims are comparative (split-trial < naive; V1 < prior reports; PFC stable over windows), please add appropriate uncertainty quantification (e.g., bootstrap CIs over sessions) and, where a difference is claimed, a test or effect size.

Reviewer #2 (Public review):

Le and Wei present a novel estimation method for information-limiting correlations. Information-limited correlations are shared noise fluctuations that affect neural encoding, but they can be hard to estimate (even to detect their presence) because they can be very small and buried under other common sources of variability that do not affect encoding. The newly proposed method bypasses two central limitations of previous approaches: extrapolation or assuming the noise structure to be Gaussian. The authors proposed a split-trial analysis where the population is split into two, and the correlations between the decoding errors arising from each population are computed. These correlations provide an unbiased measure of information-limiting correlations. The method is very simple and sound, and it is shown to deliver stable estimates with sensible magnitudes across several brain data sets. Further, even if the decoders are suboptimal, the method can detect the presence of information-limiting correlations, as only shared fluctuations of the two population decoders can possibly be observed if there are correlations that limit information.

Comments:

(1) The name "split-trial analysis" does not seem to reflect well the nature of the method introduced. I would propose something like "split-ensemble analysis" or "split-population decoding-correlation analysis".

(2) Previous work has proposed a related - but different - bootstrap method, which can be mentioned in the current paper (Nogueira et al, J of Neuroscience, 2020).

(3) The authors proposed a deconvolution method to study the shape of the distribution of information-limiting noise. An alternative would be to split neural populations into 3 or more subpopulations and compute 3rd- and 4th-order correlations between the decoding errors. This would lead to estimates of higher-order moments that can be compared to Gaussian ones and test for non-Gaussian distributions. Further, this N-split-ensemble method could be used to compare the deconvolution method results to test their consistency.

Reviewer #3 (Public review):

Summary:

In this manuscript, Le and Wei proposed a new method to identify differential correlations in real recordings (and simulations) that is based on splitting the simultaneously recorded population of neurons into two disjoint subpopulations. The method is based on evaluating the correlation between the decoded stimulus for each sub-population across trials. The authors validate their method on simulations and find the magnitude of differential correlations on three different publicly available datasets.

Strengths:

We think that this is a solid and relevant study for the computational neuroscience community, especially for the originality of the method and the fact that it seems to bypass the problem of very large populations to identify differential correlations. Overall, the results are novel and significant, and it addresses an important gap in the field. The main results are presented clearly and are easy to follow.

Weaknesses:

However, we believe that there are some additional analyses and clarifications that should be made to increase the clarity and impact of this study. In general, we believe that the authors should make a better effort to explain how their novel method depends on the number of trials and the number of neurons. More specifically:

Major

(1) The authors should show a realistic case for the covariance matrix in Figure 1. Currently, they are showing only Poisson noise (Figure 1c-e), only gain + Poisson (Figure 1f-h), and only differential correlations + Poisson (Figure 1i-k). They should show these same plots with a biologically realistic non-differential correlation structure (limited-range correlations, see Kanitscheider PNAS 2015). Perhaps even show the case for limited-range + gain + differential correlations. They should do the same for Figure 2.

(2) Throughout the manuscript, the role of population size (N) on the method is a bit confusing. Figures 1 and 2 give the impression that N is not particularly important, which is counterintuitive and surprising. We understand that that is one of the strengths of the split-trial method, but the authors should explain in much more detail in the results and methods the role of population size on their novel method. Why is large N crucial for the other methods, but not for them? There is a little bit of population-size dependency on Figures 3-5, especially on Figure 4g. The authors should explain in more detail those effects.

(3a) For dataset [27], the stimulus density was ~12 samples per deg for uniform sampling and ~1000 samples per deg for dense sampling. Figure S12 shows an overestimate of information-limiting noise when the number of trials used was significantly downsampled, which is, first of all, in disagreement with simulation results showing "when only a small number of trials are available to infer a large d-prime, split-trial analysis exhibits an under-estimation". It is true that we are not strictly in a binary classification task setting, but we are wondering if the authors have any justification for this result for [27].

(3b) Related to this point, the estimated info-limiting noise was 0.26 deg with all neurons and 0.6 deg with downsampling (we guess that is the first value of red lines in Figure S12). The only difference here, if we understand correctly, is the number of trials used. Otherwise, it's exactly the same neural responses used for estimation. So, a similar magnitude should be expected. If the latter is due to an insufficient number of trials used, would the same problem apply to the uniform sampling dataset? In other words, if there were more trials recorded with uniformly sampled stimuli, would the authors expect to see a further and significant decrease of sigma as well?

这里的 seq2seq 是基于 transformer 讲解的（前面20分钟讲解了seq2seq 的用途）

如果要看基于双RNN，得看MLDS 2018里面的讲解

基于transformer。影片2、影片3复用了2021年的课程视频

如果要看基于双RNN，得看MLDS 2018里面的讲解

actually MLDS 2018 Seq-to-seq Learning

#seq2seq * RNN with Gated Mechanism * Sequence Generation * Conditional Sequence Generation

这个作业是基于双RNN的seq2seq，而不是基于当期课程上讲的transformer

xCGidAeyS4M: ML Lecture 21-1: Recurrent Neural Network (Part I) Jan 25, 2017

2600

စာသားရဲ့ font ပုံစံကို သတ်မှတ်တာ

Halogen?

Architectural VR: Using Apple Vision Pro for Home Floor Plan Walkthroughs

• Solving the Floor Plan Scale Problem:

  • Developers Christian Selig and his partner used the Apple Vision Pro to visualize their upcoming house build in 3D, overcoming the difficulty of evaluating 2D PDF blueprints.
  • Flat floor plans fail to convey real sense of scale, room proportions, ceiling height, or sightlines, whereas virtual reality allows users to physically walk through spaces before construction.

• DIY 3D Modeling Workflow:

  • Drafting & Extrusion: Extruded 2D blueprints into basic 3D walls, floors, and ceilings using Autodesk Fusion 360 (free for hobbyists).
  • Adding Realism: Applied materials and textures (wood, stone, glass) to add visual depth and prevent spaces from feeling like generic warehouses.
  • Furniture Grounding: Imported models of real-world items (e.g., IKEA furniture, kitchen appliances) from 3D Warehouse and external catalogs to accurately judge spatial constraints.

• Custom Interactive Navigation:

  • Implemented Bluetooth controller support, terrain-following for outdoor topography, speed toggles, and height/flight adjustments between floors.
  • Added a quick real-life passthrough toggle to safely check real surroundings while moving in virtual space.

• Cost-Effective Alternative:

  • Demonstrates that high-end architectural visualization (BIM) can be achieved using accessible, free, or low-cost tools without relying on expensive enterprise architectural software.

Hacker News Discussion

• Professional Architectural Adoption (ArchViz):

  • Architects and design-build firms highlight that VR walkthroughs (using tools like Rhino3D, Revit, Enscape, and Quest headsets) are already standard in high-end projects, giving clients instant design feedback and preventing costly post-construction wall moves.

• Choice of VR Hardware & Novelty:

  • Commenters note that architectural visualization in VR has existed for over a decade on cheaper headsets (HTC Vive, Meta Quest).
  • While the Vision Pro's high screen resolution provides exceptional clarity, the core workflow relies on standard 3D export pipelines (such as USDZ/OBJ formats) rather than Vision Pro-exclusive technology.

• DIY Tooling for Homeowners:

  • Users shared alternative software choices for home modeling—such as Blender, SketchUp, and Sweet Home 3D—debating the trade-offs between precision architectural CAD and open-ended 3D modeling tools.

• Core Functionality & Flexibility:

  • The em dash (—) serves as a versatile punctuation mark capable of replacing colons, semicolons, commas, or parentheses to connect or set off related thoughts.
  • Its adaptability allows writers to mimic natural speech patterns, introduce abrupt shifts, or weave sidecar observations into a sentence without formal structural constraints.

• Differentiating Dashes:

  • Hyphen (-): Used primarily to connect compound words (e.g., "word-processor") or split words across lines.
  • En dash (–): Indicates ranges (e.g., 10–20) or joins equal-weighted compound nouns in certain formal style guides.
  • Em dash (—): Used to denote parenthetical thoughts, abrupt breaks, or strong emphasis.

• AI and Typographic Trends:

  • The em dash has become a notable marker in modern AI-generated text, as Large Language Models frequently adopt it for complex or expressive sentence structures.
  • Style guides vary on formatting: American conventions typically favor unspaced em dashes (word—word), whereas European or British conventions often prefer spaced en dashes (word – word).

Hacker News Discussion

• Pedantry vs. Practical Utility:

  • Commenters debate the distinction between hyphens, en dashes, and em dashes, pointing out that major style manuals (e.g., NYT, Chicago, BBC) frequently disagree on their exact usage rules.
  • Many argue that the em dash’s main appeal is convenience—it lets writers bypass strict rules around semicolons and colons while maintaining readability.

• AI Overuse & Writing Aesthetics:

  • Participants note that LLMs heavily overuse spaced em dashes, making their frequent appearance a key indicator of AI-generated prose.
  • Some users advocate for short, punchy sentences over complex dash-laden structures, while others feel that relying solely on brief sentences makes prose feel monotonous and overly simplistic.

• Accessibility & Keyboard Input:

  • Discussions highlight the friction of using specialized dashes in digital environments, as standard keyboards only feature a basic hyphen key, forcing reliance on auto-formatting or Unicode shortcuts.

fake photos all over network

• The Fundamental Challenge of Being:

  • Human beings inherently struggle to sit quietly without doing anything or feeling the urge to alter their current state.
  • When forced to do nothing, people experience an uncomfortable "allergy to the present moment," feeling an intense pull to fidget, check phones, or ruminate.

• Coping Mechanisms and Escape Patterns:

  • Behaviors like overeating, doomscrolling, unnecessary shopping, and starting arguments often serve as subconscious attempts to escape simple existence.
  • Research highlights how deeply people avoid being alone with their thoughts, with studies showing participants frequently preferring mild electric shocks over sitting silently in a room for a few minutes.

• Practical Exercise for Micro-Presence:

  • The author introduces a low-barrier exercise to build tolerance for existing: staying fully relaxed and open to all sensations for the duration of a single breath (inhale or exhale).
  • By using half-breath intervals (5–10 seconds) as a natural timer, individuals can practice non-defensive awareness without getting overwhelmed by formal meditation expectations.

• Benefits of Reducing Existential Restlessness:

  • Regularly practicing presence reduces the urge to constantly seek entertainment or distraction.
  • Minor daily discomforts—such as waiting in line or coping with temperature changes—become significantly easier to tolerate.

Hacker News Discussion

• Industrial Revolution vs. Task-Oriented Labor:

  • Commenters debate whether modern restlessness stems from the shift from seasonal, task-oriented work to industrial, time-based labor where hours are traded for wages.
  • Several users dispute romanticized views of historical farming, pointing out that pre-industrial agricultural life often involved relentless, backbreaking work with its own severe pressures.

• Dopamine Mechanics and "Wanting vs. Liking":

  • Discussions delve into neuroscience, noting that dopamine drives "wanting" rather than "liking."
  • Constant scrolling and fidgeting are fueled by brain circuits seeking the next stimulus, rather than genuine enjoyment of the activity.

• Practical Perspectives on Micro-Meditation:

  • Readers praise the "half-breath" framework as an approachable, low-friction entry point to mindfulness compared to rigid, multi-minute meditation routines.

• Study Overview & Core Findings:

  • MIT Sloan research evaluated LLMs on lifetime financial advice, finding that AI guidance is surprisingly sound overall.
  • AI consistently promotes positive behaviors: saving during working years, drawing down assets in retirement, investing in diversified funds, and reducing equity risk after age 45.
  • LLMs provide an accessible, low-cost alternative for individuals who cannot afford traditional financial advisors.

• Key Weaknesses & Performance Gaps:

  • AI fails to handle financial shocks well, often recommending overly drastic spending cuts after job loss even when adequate savings exist.
  • Models let investment portfolios drift passively rather than proactively recommending portfolio rebalancing.
  • Guidance frequently relies on basic rules of thumb unless given structured context.

• Prompt Quality & Demographic Disparities:

  • "Academic prompts" featuring complete financial details and explicit economic assumptions significantly improve advice quality.
  • Prompt differences linked to gender, financial literacy, and prior AI experience caused up to a 5% difference in projected retirement wealth.
  • Men and highly literate users tended to ask about strategy and growth, leading LLMs to recommend higher stock allocations, whereas women frequently included terms related to household management and family expenses.

• Implications for Financial Services:

  • AI is altering financial product discovery; LLMs frequently recommend major index providers (e.g., Vanguard, iShares) even when unprompted.
  • Financial institutions may need to optimize product visibility for LLM recommendations rather than relying solely on traditional marketing.

Hacker News Discussion

• Practical Budgeting Applications:

  • Users report positive experiences exporting local budget data (e.g., YNAB, Tiller) into LLMs to identify spending patterns, organize budget categories, and compare reward programs.

• Pitfalls in Local Tax & Jurisdictional Advice:

  • Commenters warn that LLMs struggle with location-specific tax nuances (such as city-level tax rules for S-Corp conversions) because models tend to generate immediate answers instead of asking necessary clarifying questions.

• Baseline vs. Expert Financial Advice:

  • Some participants note a Gell-Mann amnesia effect: while LLMs handle basic financial principles well, they lack depth for nuanced financial planning (e.g., sequence of return risk, asset allocation glide paths).
  • Counterarguments emphasize that generic, non-predatory LLM advice is still far superior to no advice or predatory human financial advisors charging high fees.

• Behavioral and Structural Limits:

  • Discussion highlights that key financial challenges stem from behavioral discipline or insufficient income, which high-level AI advice cannot directly solve.

explique peut-etre la gematria, si le temps n'est que le contenu des événements une durée a un sens bien plus profond

OF shill.

Worldgym

What a lovely sentiment.

We did this using the ICE tables to determine the value changes to the concentrations of the reactants and the products when the RXN heads to EQ or re-estabilishes EQ

you mention "detusche komplettsysteme" twice at least. is plenty one among them? what do you mean? i also think that distinction is not so useful. why only komplettsysteme? other competitors in general is fine imo, especially if this is a real strenght oftheirs

is this verified? i was just proposign this as a potential way to explain, not sure if its true. i mean jtl is famously focused on dach, but plepntyone? not sure if

this is too negative. first of all lets keep th 77 number everywhere for simplicty, not need to make this complex.

also when you write about the pricing, pelase start with thte costs, and then add that there may be additional because some of the marktplaetze werden von partnern gepflegt or watehver.

i mean potentially we can give more space to all the functionality that is covered here etc.

lets cut this card please

keeep rund 160, cut thte streng...

lets keep 77, for simplicty, and since they are in beta or coming soon it should be fine

produktdaten auszustpielen ist das produkt, pelase rephrase, sounds clumsy

why english? please dont

$$x^2$$ $$e^{1/2}$$

Hängt mit den terminierten Bearbeitungen und den Teamressourcen zusammen und taucht im Konzept verteilt auf. Ist wohl eher wichtig/zentral/hilfreich.

siehe Anmerkungen zu Ressourcen und Teamplanung oben! ev. geplante Bearbeitungen und zugeteilte Abteilungen matchen

Also wird eine erfasste Bearbeitung direkt im Maschinenkalender gekoppelt?

Dynamics macht wohl keinen Sinn mehr... Weil macht nicht viel - und externes Tool bringt Kompelxität. Wichtig ist, die Prüfungsdaten "revisionssicher" bzw. zertifizierungstreu zu führen.

Ich denke, dieses Modul ist ein very low hanging Früchtchen und würde mit wenig Aufwand viel Nutzen zeigen -> Schenker Freude machen bzw. motivieren

Es können mehrere Bearbeitungszeiten von div. MA erfasst werden. Jemand muss dann als "erledigt" markieren.

Insbesondere Statusänderungen sollten grafisch sehr schnell sichtbar sein (die Historie bildet ja jegwelche Änderungen ab - Instandsetzungsarbeit ausführen, Dokumente drucken, Material bestellen, Bearbeitung hinzufügen)

Allgemein Dokumente: * Wie unterschreiben? Signatur wird gemäss User eingesetzt oder wählbar? Digital signiert? * Dokumente sollten möglichst digital direkt aus der App (nachvollziehbar) versendet werden. Wenn gedruckt (Kunde wünscht das so), dann Druckdatum und Versandbestätigung. Sammeldruckjobs am Ende des Tages? * Dokumente immer on the fly produzieren oder als PDF gefixt?

Ev. ähnlich EcoLogic - ein "Bexio" mit Summen, ein "CAS-Report" mit Details...

Oder eben da Schnittstelle in Rechnungssoftware...

und in der alten separaten App - nur eben noch ziemlich unvollständig...

Heute gibt es ja die separete "WebApp" "Prüfprotokolle".

Dieses "Tool" muss sehr intelligent ausgeführt werden - alle nötigen Infos zur Priorisierung, Sicht der Abhängigkeiten etc.

Wie wird sichergestellt/überwacht, dass genügend Ressourcen vorhanden sind?

ich nehme an, das kommt öfters vor - die Bearbeitungen müssen in konkreter Anwesenheit des Teils angepasst werden. "ungeplant" wäre dann ja nur wichtig anzugeben, wenn tatsächlich Offerte existiert. Ev. nur Flag und dann bei Abrechnung einbeziehen?

Diesen Basar würde ich wenn möglich standardisieren/vereinfachen. "Rabat" ist die Hauptstadt von Marokko...

Dies können Bearb. sein, die nach Zeit und/oder Fläche abgerechnet werden oder als Fremdarbeit vergeben werden. Der Fall Fremdarbeit muss in der Terminrechnung ev. so wie "Material" behandelt werden. Z.B. Weitergegeben -> in Ausführung -> erledigt

Was passiert, wenn die Analyse verfeinert/nachgebessert wird? Muss dann eine neue Offerte generiert werden, die alte bleibt immer unabänderbar?

Nach welchen Regeln (Min-Bearbeitungsdauern, ...)?

Aber einfach auf Max-Termin? Oder werden Min-Bearbeitungszeiten, die noch zu machen sind, dazugerechnet?

Also kein Material "allgemein zu Auftrag"? und wenn, dann an eines der Hauptteile hängen?

wieso? nicht einfach auf Rechnung gleiche Reihenfolge wählen?

Mit Kamera, Scanner, als Datei?

Zubehör immer zu einem Teil (Haupt- oder Unter-) zugeordnet? Oder wie ein Teil, jedoch mit "Zubehör-Flag")

Gewisse konkrete Details müssen dann wohl nach der Anliefrung noch nachgepflegt werden!

Wo Kontrolle/QS?

können ihrerseits wieder weitere Unterteile haben -> Level "unbeschränkt"

Klarheit darüber, wie Garantieworkflows abgebildet werden sollen.

Auch andere technischen Standards, Normen etc. (die die Kundenstandards übersteuern können müssen)?

AltAppl: "Anz. identische Hauptteile" erfassbar - diese werden dann als ein virtuelles Hauptteil weiterverarbeitet (oder so ähnlich) -> Mechanismus wichtig und zu implementieren?

auch Standard-Unterteile

Werden Anlagen/Komponenten über "geklebte" QR-Codes identifizierbar gemacht?

Auch bei Kunde erfassen - also nicht nur Kontaktpersonen?

Verifizieren, ob alles Bestehende/Gewünschte sauber abgebildet werden kann und nicht "Hacks" mit "Dummy-Anlagen" gemacht werden müssen. Hängen interne Anlagen an einem spezifischen "internen Kunden" und werden ansonsten völlig analog Kundenanlagen abgewickelt?

Wahrscheinlich nur ein Thema bei neuen Kunden - ev. zu Alt-App synchronisieren für "CRM"

Rabatt-Thema über ganzen Auftrag/Rechnung klären - möglichst einfach!! %, Absolutbetrag, Positionsrabatt, Schlussrabatt, ...?

Zusätzlich noch "freie Dokumente"?

Frei hinzufügbar? Dann wäre keine nachgängige Prozessrelevanz mehr möglich (wo einblenden, wo berücksichtigen, ...)

Eine Anlage kann auch ein Gerät sein (e.g. Stapler): Ok so oder ev. anderer Name?

Nicht mehrere (rollenbasiert)?

Rollen fix oder frei erweiterbar?

Also Betreuer beim Kunde? Oder bei Schenker für den Kunden? Von wegen Popovern...

nicht nur werbeziele, generl integartions or soemthing similar aussen vor

bad writing, rephrase

super unclear, please explain this as well.....

this is super unclear, waht does this mean?

spezielle marktplaetze, instead of moecglichst viele

imo better, more accurate, what do you think?

this needs more explaination in the article. how does this number come to be and what it means. we do not want to insuniate that the company is lying, its just about comparing apples to apples here. maybe we need to write this in the intro

cut this poitn

disagree. its never about reichweite alone, the question is, waht kind of shops do they cover that others dont cover? are they foreign shops, like a strong presence in the us, are the more niche shops in some place? and therefore if you are a focused beauty brand you need this to over douglas and niche beauty? etc.??

lets use more stuff from the already publsihed articles, i am sure we have very valid pro and con points that can be used here. i mean fundamentally, and this is for the entire article, we seem to ahve not much substance in ranking the competitors. okay they have channels, some have more some have less. what about ease of use, yes pricing as well, but then what about other things you need, why a complete system is not better or is better depends on the situation. we need to write that. also the reviews might tell us something about copmelxity in use? i dont knwo, we cant make the article about have channels and pricing is this and that...describe why this tool fits this situation and the other tool fits another situation best....

so basically three points are about pricing, why? please change this

cut plaes

i would also here not make price the main part fo the article. we have with plenty and jtl two competitors that can cover everythign, and hte sales channels. and three competitors that cover only the sales channels.

so lets write it like that then. maybe we need change more int eh article than i thought...

so its either free or depends on revenue, whcih is it? also we definitely need to also point out the downside of free, the self hosting part. this is super important. we cant let the free just stand there, this is about internal resources, not about being cheaper

weak point imo

a bit too strong. yes, support can be improved, the second part can be written better, ein maechtiges werkzeug ist entsprechend komplex und benoetigt resourcen und zeit eingefuehrt zu werden.. the thrid critique seems off, i mean pricing differences by level is normal? or why is this a point? 4 do we know that for sure, i rather kick this point or we combine this with point 3 to summarise something about pricing...

dont like the how its said, fuehrt den betrieb selbst? sounds off to me. and also the part of own shop auf einer datenbasis? auf einer datenbasis, what does that even mean

lets soften this because its hard to get this right and keep it up to date, keep it more generic please

this is very unclear, what are you trying to say? which partner, which anbieter? so its partners of plentyone making the connections? or it costs more for runnign the connections?

am counting 93 or so on this page https://www.plentyone.com/marketplaces?_gl=117fpmvj_upMQ.._gaOTQ4MTM5ODQxLjE3ODU2OTE2MDY._ga_FMMRRE5Y44*czE3ODU2OTE2MDUkbzEkZzAkdDE3ODU2OTE2MDUkajYwJGwwJGgyMDg1NTcyNDg.

not sure, what do you think? is this a different number

not very prcise. the difference seems clear, but you wrotie it in a suboptimal way

welches raster? nicht gut definiert, keine gute ueberleitung.

The author explains that economic sanctions serve political purposes instead of economic purposes, containing multiple financial and trade limitations. Sanctions will interfere with global logistics and international trading supply chains, raising transportation and import costs for overseas goods.

Sanctions vary widely in duration and coverage, and long-term comprehensive ones evolve into economic warfare.

This raises an interesting question: must they?

Suppose instead of patterns or patterns-turned-to-components (libraries/dependencies) that Arnout lays out, we instead took an approach to component re-use where packages are authored rather to be dev dependencies (in NPM parlance) rather than ordinary dependencies. Modules would neither be dynamically loaded nor statically treeshaken-and-bundled, but rather a module/class would be annotated with an interface, and the dev dependency would (a) carry the implementation payload and (b) be responsible for analyzing the classes that are expecting to implement such pattern and then modify the source files of a conforming consumer in-place to provide the necessary method/property implementations.

The observer pattern would be a good use case.