The common adolescent bipolar phenotype shows positive biases in emotional processing


  • GMG has held grants from Sanofi-aventis and Servier; has received honoraria for speaking engagements from AstraZeneca, Bristol-Myers Squibb, Eisai, Lundbeck, Sanofi-aventis, and Servier; and has served as a consultant to AstraZeneca, Bristol-Myers Squibb, Eli Lilly & Co., Lundbeck, P1vital, Sanofi-aventis, Servier, and Wyeth. CJH is on the advisory board of P1vital and owns shares in the same company; and has received consultancy fees from Servier, GlaxoSmithKline, AstraZeneca, Lundbeck, Merck, Sharpe and Dohme, and P1vital. PLR has no conflicts of interest to report.

Corresponding author:
Dr. Catherine J. Harmer
University Department of Psychiatry
Warneford Hospital
Oxford OX3 7JX, UK
Fax: 44 (0)1865 251076


Rock PL, Goodwin GM, Harmer CJ. The common adolescent bipolar phenotype shows positive biases in emotional processing.
Bipolar Disord 2010: 12: 606–615. © 2010 The Authors.
Journal compilation © 2010 John Wiley & Sons A/S.

Objectives:  Bipolar disorder is associated with abnormalities in emotional processing that persist into periods of remission. However, studies of euthymic bipolar disorder patients may be confounded by the experience of mood episodes and medication. We therefore assessed an adolescent group for vulnerability markers associated with the bipolar phenotype.

Methods:  The Mood Disorder Questionnaire (MDQ) is a screening tool for bipolar disorder that targets mood-elevation symptoms. We selected 32 high-scoring students (≥ 7 symptoms) with the adolescent bipolar phenotype and 30 low-scoring controls (≤ 3 symptoms) and screened them with the Mini International Neuropsychiatric Interview–Plus for bipolar disorder and other psychiatric disorders. We investigated emotional processing by assessing facial expression recognition, emotional memory, emotion-potentiated startle, and a dot-probe task.

Results:  Of the high-MDQ participants, 12 were in remission from bipolar disorder defined by DSM-IV-TR and interview (bipolar II disorder/bipolar disorder not otherwise specified) and 3 from major depressive disorder. High-MDQ participants had higher levels of neuroticism, low mood, and lifetime anxiety comorbidity and alcohol dependence compared with low-MDQ participants. The high-MDQ group showed facilitated recognition of surprised and neutral facial expressions and enhanced processing of positive versus negative information in emotional recognition memory and emotion-potentiated startle. There were no effects on emotional categorisation/recall memory or attentional bias in the dot-probe task.

Conclusions:  These results suggest that students with the common adolescent bipolar phenotype show positive emotional processing biases despite increased levels of neuroticism, low mood, and anxiety. Such effects may represent a psychological vulnerability marker associated with the bipolar phenotype.