On 2013 Dec 30, Ben A Inglis commented:

It seems to me that although the intent was to assess connectivity induced by reading a novel, in the absence of a control task there is no way to know whether the changes were due to the specific act of reading or due simply to the subjects' being engaged in a structured task over the study period. Surely, to be confident that the changes are due to reading, a second group of subjects should have been exposed to a control task that was as similar as possible to the novel reading. For example, control subjects could have been instructed to watch a dramatic TV mini-series, with each evening's exercise lasting approximately the same time as the reading. (Questioning of control subjects could have verified that the task was completed as instructed, in precisely the same manner as the test subjects had their reading task assessed.) An even better control would have been to use the same story but change the medium. For instance, the test subjects might have read the story while the control subjects listened to it on an audio book. Perhaps there is even need for a second control group; subjects are set a task to attend to each night, but there is no plot to follow. In sum, without assessing directly the effects of a structured attention task over the study period there is no way to be sure that reading per se is causing the connectivity changes.

On 2013 Dec 30, Ben A Inglis commented:

It seems to me that although the intent was to assess connectivity induced by reading a novel, in the absence of a control task there is no way to know whether the changes were due to the specific act of reading or due simply to the subjects' being engaged in a structured task over the study period. Surely, to be confident that the changes are due to reading, a second group of subjects should have been exposed to a control task that was as similar as possible to the novel reading. For example, control subjects could have been instructed to watch a dramatic TV mini-series, with each evening's exercise lasting approximately the same time as the reading. (Questioning of control subjects could have verified that the task was completed as instructed, in precisely the same manner as the test subjects had their reading task assessed.) An even better control would have been to use the same story but change the medium. For instance, the test subjects might have read the story while the control subjects listened to it on an audio book. Perhaps there is even need for a second control group; subjects are set a task to attend to each night, but there is no plot to follow. In sum, without assessing directly the effects of a structured attention task over the study period there is no way to be sure that reading per se is causing the connectivity changes.