Examining Affective Forecasting Biases in Adolescents with Depressive Symptoms: An Ecological Momentary Assessment Approach

Anna Tierney, BA

McLean Hospital – Research Assistant
Tierney_Anna poster

Scientific Abstract

Background: Beck’s cognitive theory posits that individuals with depression hold inaccurate, negative views about themselves, the world, and the future, including predictions of future mood (Beck et al., 1979). Initial research suggests that adults with elevated depressive symptoms may indeed have pessimistic predictions of their future mood (Wenze, Gunthert, & German, 2012; Zetsche, Bürkner, & Babette, 2019). In contrast, other research suggests that depressed adults have lower optimistic biases relative to non-depressed adults (i.e., are more accurate in their predictions, also known as “depressive realism”) (Alloy & Abramson, 1988). However, these studies used brief time periods (e.g., only 1 week) to gauge predictions and relied on adult populations. The present study used repeated daily ecological momentary assessment (EMA) over 1 month to examine the extent to which adolescents with varying levels of depression are accurate in their prediction of daily positive affect (PA) and negative affect (NA).

Methods: Participants included 64 adolescents aged 12 – 18 years with varying levels of depressive symptoms (Center for Epidemiological Studies Depression Scale; CES-D; Radloff, 1977) who reported predictions of their future affective states. They then completed 30 days of smartphone-delivered EMA (2-3 surveys per day) to assess their PA and NA (Forbes et al., 2012). 

Results: Participants’ guesses of their PA and NA were significantly higher than their actual (EMA) mean PA (t(63) = 5.73,  < .001) and NA (t(62) = 3.44,  = .001). In addition, higher levels of depressive symptoms were significantly associated with a greater NA bias (t(54) = 3.03,  = .004) but not PA bias (t(54) = -0.23,  = .82), even when adjusting for mean levels of NA and PA. We are currently collecting a separate, independent sample (planned n = 148) to test the generalizability of these findings.

Conclusion: In contrast to the “depressive realism” hypothesis, and consistent with Beck’s cognitive theory, participants with relatively higher depressive symptoms exhibited pessimistic affective forecasting bias, such that they predicted higher levels of future NA than what they experienced.

Live Zoom Session – March 9th

research Areas


Anna O. Tierney, BA, Hallie A. Brown, BA, Emma G. Balkind, BS, Aruni S. Ahilan, Brianna M. Pastro, BS, Elana S. Israel, BA, Erika E. Forbes, PhD, Diego A. Pizzagalli, PhD, Christian A. Webb, PhD

Principal Investigator

Christian A. Webb, PhD