Category: Assessment

PS6- #C67 - Ability to Recall Autobiographical Memories: Psychometric Properties of the Autobiographical Memory Test (AMT)

Friday, Nov 17
2:45 PM – 3:45 PM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Psychometrics | Cognitive Vulnerability | African Americans/Black Americans

The Autobiographical Memory Test (AMT) assesses overgenerality in memory by presenting a set of positive and negative cue words and instructing participants to recall an autobiographical event related to each word that is specific in time and place. Overgenerality consists of responding to cues with memories that are categories or summaries of similar events rather than specific events. As a cognitive style, overgenerality constitutes a risk factor for the development, maintenance, and exacerbation of depression, as well as suicidality, trauma, problem-solving difficulties, and impaired ability to imagine the future.


As might be expected, the focus of research using the AMT has been on the overgenerality and/or specificity of responses. However, not commonly discussed is that not infrequently participants provide responses that are neither specific nor overgeneral, but reflect a failure or inability to recall an autobiographical memory. These responses consist of a comment on the cue word (e.g., “I like being happy”, coded as “semantic associate”) or a report of not having a memory (coded as “omission”). Given that certain populations may exhibit elevated rates of “failure to perform the task” (Beyderman & Young, 2014), this phenomenon within the AMT may be significant in itself.


The current study conducted an IRT analysis of the AMT along the latent trait of “ability to perform the task” of autobiographical memory retrieval. Data came from a sample of 100 African American psychiatry outpatients, 36% of whom met the diagnostic criteria for major depressive disorder.  Difficulty parameter values of the test items ranged from -3.57 to -0.69; M(SD)=-2.32(2.34), with an exception of an outlier -11.37. Item difficulties were below the mean level (i.e., 0) suggesting that it was relatively easy for participants to perform the task of the AMT. Almost all items had substantial levels of discrimination: range 0.31-2.37; M(SD)=1.26(0.51), suggesting that the cues had adequate ability to discriminate between different levels of the latent trait of “ability to perform the task.” The greatest test information was between -2.5 and -1.0, indicating that the version of the AMT employed is more sensitive in detecting “ability to perform the task” at lower levels of this trait, which pertains to groups of most clinical interest.


These findings have potential clinical implications. Given the importance of healthy autobiographical memory functioning for psychosocial adjustment and well-being, a measure with solid psychometric properties targeting identification of individuals with impairments in autobiographical memory retrieval is strongly needed. The autobiographical memory retrieval score also may serve as a validity index for the OGM score, since overgenerality of retrieved memories cannot be assessed in the context of a deficit in retrieving memories. Once identified, difficulties in autobiographical memory retrieval, which may reflect phenomenon such as unresolved trauma, may become foci of interventions and inform the choice of treatment (e.g., Cognitive Processing Therapy).  

Irina Beyderman

doctoral candidate
Illinois Institute of Technology (IIT)
Northbrook, Illinois

Michael Young

Professor of Psychology
Illinois Institute of Technology
Chicago, Illinois