Revisiting Depression's Cognitive Downside

Depression, by h.koppdelaney

Is depression actually good for you?

Experts now believe that mild to moderate depression may be good for us – and even help us live longer. Rebecca Hardy explains how to reap the benefits

We constantly hear how depression is blighting our lives, but some experts have an interesting, if controversial, theory: depression can be "good for us", or at least a force for good in our lives.

Is this the start of a new Negative Psychology¹ movement? Let's all seek out personal tragedy, sadness, insomnia, and a profound sense of failure and hopelessness, because it's good for us!!


Last year, author and blogger Jonah Lehrer had a lengthy (and controversial) essay in the New York Times Magazine on Depression's Upside. The main idea, that depression has cognitive and evolutionary advantages, was largely based on a review paper by Andrews and Thomson (2009). In it, they put forth the analytical rumination hypothesis: depression is an evolved response to complex problems, and focusing on them to the exclusion of everything else is beneficial.

In response, The Neurocritic was motivated to write about Depression's Cognitive Downside:

On the contrary, numerous papers have shown that impairments in cognitive processes such as executive control, attention, and memory persist after a depressed person has recovered (Andersson et al., 2010; Baune et al., 2010; Hammar et al., 2009). In actively depressed patients, Baune and colleagues (2010) found impairments in all domains tested: immediate memory, visuospatial construction, language, attention, and delayed memory. These deficits can contribute to lower social and occupational functioning and a diminished quality of life. In addition, depression can be associated with declines in problem solving abilities on neuropsychological tests such as the Wisconsin Card Sorting Test and the Tower of London test.

Now, a new paper by von Helversen et al. (2011) has claimed that depression is good for decision making. Lehrer wrote about this study as support for the analytical rumination hypothesis in Does Depression Help Us Think Better?

Here’s where things get interesting: depressed patients approximated the optimal strategy [for hiring the best applicant in a simulated job search] much more closely than non-depressed participants did. The main problem with healthy subjects is that they proved lazy, unwilling to search through enough applicants. Those with depression, on the other hand, were much more willing to keep on considering alternatives, which is why they performed far better on the task. While this study comes with many caveats, it remains an interesting demonstration that depression, at least in specific situations, seems to enhance our analytical skills, making us better at focusing on social dilemmas.

Participants in the study were 37 inpatients diagnosed with major depression upon admission to the hospital (10 of whom were omitted "due to technical difficulties with the choice task"). The 27 remaining patients were classified as either "depressed" (n=15) or "recovered" (n=12) based on improved scores on the Patient Health Questionnaire (PHQ-D) between admission and testing (which was a mean of 6.25 days -- that seems like an incredibly rapid remission, which makes one wonder about the actual severity and why they were admitted in the first place). Only half of the patients, both depressed and recovered, were on antidepressants (none were on other medications), which seems unusual for patients who may have been suicidal. Perhaps the criteria for admission to the psych ward in Germany are different than they are in the U.S. and Canada. The still-depressed patients were in hospital an average of 4.20 days when they were tested (which was not significantly different from the recovered patients). It wasn't completely clear if any of the patients were already on antidepressants, or whether the pharmacological treatment started during hospitalization for those on meds.² The paper did not state whether any of the depressed patients had another diagnosis, such as an anxiety disorder of any sort (co-morbidity is common).

Mean scores on the Beck Depression Inventory (BDI) were higher in the Depressed group (29.13) than in the Recovered group (16.67) or the Control participants (6.63), who also differed from each other. BDI scores of 14–19 are considered mildly depressed, 20–28 moderately depressed, and 29–63 severely depressed. So patients in the Depressed group scored at the low end of severely depressed, the Recovered participants were mildly depressed, and the Controls (n=27) were not depressed at all.

The task administered to all participants is called the "Secretary Problem":

The sequential decision-making task consisted of playing 30 games of a secretary-type problem. Each game challenged participants to find the best candidate for a job out of a sequence of 40 applicants. The 40 applicants were presented one after another, in a random sequence. After an applicant was presented, participants needed to decide whether they would accept the applicant or not. If they accepted the applicant, the game concluded and the next game started. If they rejected the applicant, the next applicant was presented. Rejected applicants could not be chosen later in that game.

Information about the current candidate included their relative ranking compared to the candidates that came before, but not their absolute ranking. Points were awarded based on the absolute ranking of the candidate chosen on each round. If a participant didn't make a choice until the end of the sequence, they were forced to accept the final candidate. So it seems that an indecisive person would be more likely to continue the search for a longer time...

Results showed there was a trend in that direction (p=.08): search length was 23.37 for Depressed, 16.87 for Recovered, and 17.96 for Controls. Performance goals for each round (how good a candidate would have to be in order to be chosen) and the relative rank of candidates did not differ between groups. However, the number of points awarded for each game did differ (p=.02): 37.67 for Depressed, 35.50 for Recovered, and 35.17 for Controls. A computational model suggested that the Depressed group had higher internal thresholds for the first and second, but not the third threshold. A caveat from the authors:

However, although we found that depressed participants had higher thresholds than did nondepressed participants, we did not find significant differences in the self-reported goals of participants. This suggests that differences in behavior may not result from participants’ conscious effort to perform well. Thus, increases in thresholds could be an artifact stemming from greater persistence and the inability to disengage from a task.

What does this mean? That severely depressed inpatients should be given the task of selecting job candidates for Fortune 500 companies, while they are so impaired otherwise that they are unable to work or function socially? Is a very modest performance benefit in a laboratory sequential decision making task worth the pain and suffering of severe depression, along with its concomitant deficits in other cognitive domains?


Footnotes

¹ This is opposed to the Positive Psychology movement.

² Any "recovery" from depression within six days has nothing to do with the start of antidepressants, unless it's a placebo effect. Treatment effects are generally not seen for 4-6 weeks.

References

Andersson S, Lövdahl H, Malt UF. (2010). Neuropsychological function in unmedicated recurrent brief depression. J Affect Disord. Jan 18. [Epub ahead of print]

Andrews PW, Thomson JA Jr. (2009). The bright side of being blue: depression as an adaptation for analyzing complex problems. Psychol Rev. 116:620-54.

Baune, B., Miller, R., McAfoose, J., Johnson, M., Quirk, F., & Mitchell, D. (2010). The role of cognitive impairment in general functioning in major depression. Psychiatry Research 176:183-9.

Hammar A, Sørensen L, Ardal G, Oedegaard KJ, Kroken R, Roness A, Lund A. (2009). Enduring cognitive dysfunction in unipolar major depression: A test-retest study using the Stroop paradigm. Scand J Psychol. 2009 Dec 23.

von Helversen, B., Wilke, A., Johnson, T., Schmid, G., & Klapp, B. (2011). Performance benefits of depression: Sequential decision making in a healthy sample and a clinically depressed sample. Journal of Abnormal Psychology DOI: 10.1037/a0023238