Divided attention at encoding or retrieval interferes with emotionally enhanced memory for words

ABSTRACT

Emotional information is typically better remembered than neutral information. We asked whether emotional, compared to neutral, words were less vulnerable to the detrimental effects of divided attention. In two experiments, undergraduate students intentionally encoded words of intermixed valence (neutral, negative, or positive) and arousal (neutral, high, or low). Following a filled delay, memory was assessed with a free recall test. In Experiment 1, participants encoded visually-presented words under either full attention (FA; no distracting task) or divided attention (DA; concurrently making animacy decisions to auditorily-presented distractor words) in a counterbalanced, within-subjects design. As expected following FA at encoding, recall was significantly enhanced for negative compared to neutral words. Following DA at encoding, recall was significantly impaired across all valences. Critically, DA at encoding also eliminated the memory benefit for negative information: recall of negative words was no longer significantly different from neutral or positive words. In Experiment 2, we manipulated attention at retrieval rather than encoding. Remarkably, results from Experiment 1 were replicated: DA eliminated the well-known emotionality boost for negative words. In both experiments, memory for positive words did not significantly differ from neutral. Findings suggest that DA during either encoding or retrieval can interfere with the specific mechanisms by which negative emotion typically improves memory.

KEYWORDS:

• memory
• emotion
• emotionally enhanced memory
• divided attention

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Data that support the findings of this study are openly available on the Open Science Framework at https://doi.org/10.17605/OSF.IO/4HZMU.

Notes

1 The pattern of results reported below remained the same with the full sample included.

2 As in Experiment 1, the pattern of results reported below remained the same with the full sample included.

Additional information

Funding

This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) through an Alexander Graham Bell Canada Graduate Scholarship (CGSD3-535024-2019) awarded to author RCY, and an NSERC Discovery grant (2020-03917) awarded to author MAF.