Research Article
Exploring Affect Recall Bias and the Impact of Mild Depressive Symptoms: An Ecological Momentary Study
@INPROCEEDINGS{10.1007/978-3-030-25872-6_17, author={Desir\^{e}e Colombo and Carlos Suso-Ribera and Javier Fernandez-\^{A}lvarez and Isabel Felipe and Pietro Cipresso and Azucena Palacios and Giuseppe Riva and Cristina Botella}, title={Exploring Affect Recall Bias and the Impact of Mild Depressive Symptoms: An Ecological Momentary Study}, proceedings={Pervasive Computing Paradigms for Mental Health. 9th International Conference, MindCare 2019, Buenos Aires, Argentina, April 23--24, 2019, Proceedings}, proceedings_a={MINDCARE}, year={2019}, month={7}, keywords={Ecological momentary assessment Momentary affect Recall bias}, doi={10.1007/978-3-030-25872-6_17} }- Desirée Colombo
Carlos Suso-Ribera
Javier Fernandez-Álvarez
Isabel Felipe
Pietro Cipresso
Azucena Palacios
Giuseppe Riva
Cristina Botella
Year: 2019
Exploring Affect Recall Bias and the Impact of Mild Depressive Symptoms: An Ecological Momentary Study
MINDCARE
Springer
DOI: 10.1007/978-3-030-25872-6_17
Abstract
Traditional clinical and research assessments rely on retrospective questionnaires, that ask individuals to retrospectively summarize how they felt during the last period. Nevertheless, people are not accurate at recalling past experiences without altering the content, especially when they are required to report their affect. In this study, we adopted a smartphone-based ecological momentary assessment (EMA) to collect daily assessments of positive (PA) and negative (NA) affect throughout two weeks in a sample of healthy students (n = 47). Results showed that both PA and NA are subject to the recall bias; more specifically, people tended to overestimate both affects during the retrospective assessment. This bias was influenced by the presence of mild depressive symptoms as measured by the Beck Depression Inventory (BDI), which led participants to a greater overestimation of NA and higher underestimation of PA. While NA bias was more context-dependent, PA bias showed more stability across time.