The impact of stress and emotion on item-method directed forgetting

Abstract

Directed forgetting is a classical paradigm in cognitive psychology that serves to investigate intentional memory control processes. It illustrates the observation, that mnemonicinformation can be forgotten deliberately and on demand. Of the two common variants of this paradigm – the list-method and the item-method – this dissertation uses the latter. In item-method directed forgetting, stimuli are presented one at a time, each followed by a cue to either forget (F) or remember (R) the preceding stimulus. In a final test (such as free recall of recognition), memory is generally better for R stimuli as compared to F stimuli. The effect has been attributed mostly to two underlying processes: On the one hand, stimuli are ‘selectively rehearsed’ after an R cue. On the other hand, relatively recent findings from neurophysiological research suggest that stimuli are ‘actively suppressed’ after having received an F cue.The present work contains four studies that investigate if, to what extent, and how this memory phenomenon is affected by different forms of stress and emotion. This research question is based on the clinical observation, that posttraumatic stress disorder (PTSD) patients especially suffer from emotionally and physically intense, traumatic recollections (i.e., intrusions), which are completely out of their control. This, accordingly,leads to the question whether people with PTSD are basically able to deliberately control their memories and which factors contribute to this process.In Study I, we examined this question and conducted an item-method directed forgetting experiment in a sample of Ugandan civil war refugees. All of them had experienced massive traumata, about 50% of the sample fulfilled PTSD criteria. Like in all of the following studies, we used complex photographs (depicting motives from participants’ everyday life) as stimuli.Memory was tested by recognition. All of the stimuli presented in learning trial (i.e., F- and R stimuli) were shown to the participants once again but this time with intersperseddistractor pictures that were very similar to the original stimuli. Directed forgetting was found in the non-PTSD group but not in the PTSD group. Moreover, there was a correlation between the individual directed forgetting effect and the average arousalrating that the respective person had assigned to the stimuli. The findings raised the question whether the effect is more due to characteristics of the person (i.e. having experiencedtraumatic stress) or due to stimulus characteristics (e.g., valence, arousal).We therefore investigated in study II how experimentally induced stress by means of the Trier Social Stress Test (TSST) affects directed forgetting of stimuli that differ in valence.Stress - even if not as intense as traumatic stress - is known to affect memory consolidation and retrieval on different levels. If, however, intentional memory control is affected by it, has not been investigated before. Whereas the stress condition did notinfluence directed forgetting, an effect of picture valence (and, implicitly, arousal) was found. Neutral pictures were forgotten, positive ones, however, were not.Whether this finding generalizes to the full spectrum of emotional (pictorial) material was then examined in study III. We varied picture valence and arousal systematically and could further substantiate our suspicion that directed forgetting is shown for low- but not for high-arousing stimuli. Interestingly, both in study I and study II the reduction of directed forgettingwas mostly due to false alarms (i.e. new stimuli that are identified as old ones). When only hits (i.e. old stimuli that are correctly classified as old) were taken into account, however, directed forgetting could be observed which contradicts deficits in the processes (selective rehearsal, active inhibition) thought to underlie the effect.In study IV, we therefore raise the question whether in item-method directed forgetting the discrepancy between hits and false alarms patterns might be explained by another processinvolved (beyond the mechanisms mentioned above), such as an ‘alert(ing)’ induced by the R and F cues (defining ‘alert’ as more stimulus-bound and ‘alerting’ as the organism’s reaction to it). Research on thought suppression has repeatedly demonstrated socalled ‘ironic processes’, which refers to the fact that material, that is to be ignored, is actually rendered more prominent. In order to investigate the effect of a postulatedironic alerting, we introduced another instruction without any cue (U for ‘uncued’) that was presented as often as R and F cues during learning. Our data indicate that such an ironic effects also takes place in directed forgetting: F stimuli were recognized worse than R stimuli (directed forgetting) but also better than U stimuli (ironic effect).The present dissertation demonstrates that stress per se does not reduce item-method directed forgetting (person characteristic). However, data suggest that the effect is modulated by the stimulus material and its ‘arousing potential’ (stimulus characteristic). In healthy participants, stimuli had to be high-arousing to impair dierected forgetting (studies II, III). PTSD patients, however, process - because of an altered (i.e., ‘overconnected’ and hyperaroused) stimulus perception - even material that is rated by healthy individuals as neutral as highly emotional and personally relevant which leads to the same result (study I). The conclusion that, in PTSD, this is mostly due to a lack of inhibition of F stimuli is challenged by study IV. Lack of inhibtion may still be an important factor but there should also be considered the possibility of an overshooting alert(ing) in response to R and especially F cues.