-
Reviews
Back to Reviews
Submited on: 23 Sep 2010 02:22:07 PM GMT
Published on: 23 Sep 2010 04:37:51 PM GMT
- Other Comments: The reported research is interesting. The modulation of cognition by mood is not sufficiently appreciated, and so such research is desirable. However, the current report contains some flaws that significantly limit interpretation of results. The main problem is that the experiment is confounded by the methods used to induce positive and negative mood. The authors employed an independent groups design. Ideally the procedure for the two groups would vary by a single factor, but this was not the case. The positive mood induction was via a film clip and the negative mood induction by reading. This variable, reading/watching is a confounding variable. It could easily be argued that the experiment was about the influence of reading text verses watching TV on unintentional learning. This significantly weakens interpretation of results as you can no longer infer cause and effect; rather you have a form of correlation between performance and mood state. The task used to measure unintentional learning appears to have been rather difficult, was this intentional? As performance in both groups appears to be not much better than chance, the authors should have considered performing a signal detection analysis, this would avoid problems of response bias that probably influence their results. In the reporting of the mood induction, the authors fail to provide us with the mean scores of the two groups prior to mood induction. Although we could expect these two figures to be very similar considering the group sizes, they should be presented. Indeed the report would benefit from a table or better still a graph showing the mood ratings of the two groups at the three time points. To show that the mood induction was effective the correct ANOVA to report would have had all three levels of time point, not just two. Without further information, the validation of the mood induction paradigm is not justified. There are also some wording and format issues. A large proportion of the discussion section involves a psychodynamic explanation. This should be mentioned in the abstract. At several points the authors talk about not finding an effect in the sad group; however their design was not able to do this anyway. Only in a repeated measures design can you say things about a single group, in an independent groups design such as this you can only examine the effects between groups. Overall this research adds something to the literature, but suffers from faults that limit interpretation. It could be improved somewhat by the addition of additional data and analysis as described above.
-
Invited by the author to review this article? :
No -
Have you previously published on this or a similar topic?:
No
-
References:
None -
Experience and credentials in the specific area of science:
None - How to cite: Pluck G .Issues with method and analyses[Review of the article 'Mood Induction and Unintentional Learning in Healthy Subjects ' by Kwok S].WebmedCentral 2010;1(10):WMCRW0049
Main claim: Mood induction can depend on types of learning and, in particular, sadness induction is not shown to improve the ability of unintentional learning of negative images.
This claim is important to better understand the effects of mood induction and how they vary according to the task at hand.
Yes
Yes
Yes
No
Yes
Cf other comments
It is an important work I would enjoy discussing at a laboratory seminar. A follow up study including all the suggestions mentioned in section "other comments" could lead to an outstanding paper.
Dr. Kwok proposed a novel mood induction procedure on 51 healthy subjects to test if subjects with induced sadness would exhibit better unintentional-learning abilities of negatively charged images when compared with happiness-induced subjects. He found that subjects with induced happiness had significantly better unintentional-learning abilities of negative images and concluded there was no evidence for mood congruence in subjects with induced sadness.
The study of Dr. Kwok is particularly interesting for its novel design to test mood congruence effect in unintentional learning. In particular, the protocol was demonstrated to induce lasting mood differences between the two groups studied. The results are of special interest as they suggest that mood congruence may not be present in unintentional learning, and that the mood congruence findings may be depending on the learning processes at stake.
It would have been informative to add the results of the t-test comparing performances on positively charged images. Could it be that sadness-induced subjects remembered worse overall? Also, the mood induction was shown to last the whole experiment and scores to stabilize towards the end for both groups around an average score of ~4.5. This suggests that this score corresponds to a “neutral mood”. Considering this was also the score immediately after mood induction for the sadness-induced subjects, it could be that the positive mood induction was much more effective than the negative mood induction. The addition of a third “neutral group”, while very demanding experimentally, would be an ideal solution to investigate this potential issue. The sample size was satisfactory, and more statistical power could have been added by accounting for the memory ability of each participant. The experiment could also benefit from showing a sad scene from the same movie, or similar movie, rather than having a different task (reading).
In terms of presented data, it would have been beneficial to indicate, in addition of the mean, the 95% confidence interval or standard deviation of the group correct-identification scores. The mean and standard deviation of the group mood scores, at each mood evaluation, would be useful as well.
In conclusion, Dr. Kwok presented a novel and useful mood induction procedure which also proved efficient through its lasting effect during the whole experiment. His results are an important contribution to the investigation of mood congruence effects on unintentional learning. A follow-up study better assessing the state of sadness, with more comparable mood induction tasks, and possibly introducing a third “neutral” group, could provide even more decisive results to this compelling work.
Yes
No
None
I have work on fMRI studies which adopt similar investigations and with similar protocols.