Mood states effect retrieval not encoding in item-method directed forgetting

ABSTRACT

Three experiments examined the influence of mood state on item-method directed forgetting. In Experiments 1 and 2, happy, sad or neutral mood states were induced before the study or in Experiment 3 before retrieval. There were no significant effects of mood on memory in Experiments 1 or directed forgetting costs and benefits in Experiment 2. In Experiment 3, mood states influenced costs and benefits of directed forgetting among happy and neutral participants and costs alone among sad participants. There were no changes in the overall recall of remember-cued items when mood states were induced before or after the study. Explanations for the effect of mood on the retrieval of remember- or forget-cued items are discussed. Moods dissipate when induced prior to encoding, but appear to influence memory when induced prior to retrieval.

KEYWORDS:

• Item-method directed forgetting
• costs and benefits of directed forgetting
• mood states

Acknowledgment

We thank Dr Sue Sherman and two anonymous reviewers for their helpful comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the author’s.

Notes

1 In Experiments 1–3, recall was reanalysed taking into account whether participants correctly attributed the item to being a remember- or a forget-cued item at study. Results did not differ from that reported above in each of the three Experiments when source attributions were taken into account.

2 While the design for Experiments 2 and 3 included a Remember-all group, the outcome of interest in Experiments 2 and 3 was similar to that of Experiment 1, that is, The Mood × Instruction interaction. Consequently, the same power analysis in Experiment 1 was used in Experiments 2 and 3. More participants were tested in Experiments 2 and 3 than what was necessary to obtain a power of 0.80 because participants were available, and we wanted to take advantage of that opportunity to detect a potentially subtle three-way Group × Mood × Instruction interaction. This power analysis detects medium to large effect sizes, it may not detect small effect sizes.

3 A 2 Group (Directed Forgetting, Remember-All) × Mood (Happy, Neutral, Sad) analysis of variance on the composite arousal measure indicated only a Mood effect F(1,104) = 5.23, MSE = 0.772, p = 0.007, η² = 0.091. Bonferroni multiple comparison tests indicated that sad participants were more aroused (M = 2.56, SD = 0.85) than Happy (M = 1.95, SD = 0.82), and neutral (M = 1.96, SD = 0.98, p’s = 0.025, 0.014) who were not significantly different.

4 Gender effects were included in this analysis at the suggestion of an anonymous reviewer.

5 A 3 (Mood: Happy, Neutral, Sad) × 2 (Group: Directed-forgetting, Remember-all) × 2 (Instruction: Remember, Forget) analysis of variance on the composite arousal scores showed significant Mood effect F(2,101) = 22.196; MSE = 0.538, p = 0.00001, η² = 0.305). Sad participants were more aroused (M = 2.66, SD = 0.90) than Happy (M = 1.51, SD = 0.45, p = 0.0001) and Neutral participants (M = 1.95, SD = 0.79, p = .0001) who also differed in their level of arousal, p = 0.0001. Since arousal level was not qualified by any interaction effect in Experiment 2 or 3, arousal effects do not alter results or discussion presented in either Experiment. Arousal effects will not be discussed further.