T3. Retrieval goal and forgetting

Concerning the goal-related relevance explanation of directed forgetting, we can assume when people observe another person with the same intention to learn, and see that this person is instructed to forget previously studied information, then they will produce the same intentional forgetting effect as the person they observed. This seems to be an important aspect of human learning: if we can understand the goal of an observed person and this is in

line with our behavioural goals then our learning performance will mirror the learning performance of the model. In a series of directed forgetting experiments we investigated the relationship between the goal of the participants and the effect of forget instruction.

However, in a standard list-method directed forgetting procedure it is difficult to independently manipulate the goal of the participants and the type of the instruction they receive, as they are strongly associated in this paradigm, so it was detailed in Racsmány et al.

(2012) the goal of the participants and the instruction they receive is hardly dissociable in the standard DF procedure. A possible way to circumvent of this problem if participants are not directly instructed, but observe another person, who receive either a ‘forget’ or a ‘remember’

instruction. This was done in 4 experiments using a modified directed forgetting procedure.

In Experiment 1 and 2 participants watched a movie of a directed forgetting experiments. They either receive a ‘simple observation’ or a ‘goal sharing’ instruction, the latter instruction make the participants to share the learning goal of the actor in the movie. In Experiment 3 and 4 followed the same logic than the previous two experiments with the exception that this time the observed model was a real participant in the experiment. The results were straightforward, participants produced directed forgetting effect only with the goal sharing instruction, and not with the simple observation instruction. In Experiment 3 and 4 the recall performance of the observer mirrored the performance of the model. Our results support the assumption that suppression of episodic memories is not automatically generated by environmental cues but depends on the goals of the person who encodes and retrieves them.

This assumption is further supported by still unpublished findings from a recent study of my laboratory. In this study we aimed to investigate the role of post-instruction encoding and pre-instruction relevant information in successful directed forgetting. Using a 2-list directed forgetting procedure and designated the second list as to-be-forgotten items (in Experiment 3) we were able to formulate different hypotheses based on the retrieval inhibition and context-change accounts. Retrieval inhibition theory assumes that forgetting of the to-be-forgotten items serves the adaptive goal of the participants to facilitate the learning of relevant information throughout the experiment (Bjork, 1989). Based on this assumption, we should not assume that the relevant information must follow the instruction. The F-instruction designates the to-be-forgotten items as irrelevant information, however, the participants need to learn some relevant information in order to being an adaptive behavior

149

disregarding irrelevant information. From this point of view, it is not necessary to assume that the relevant information should follow the F-instruction, the only requirement for them is to be encoded before the final recall of all studied items. The reset-of-encoding hypothesis (Pastötter & Bäuml, 2010) would hypothesize that using two study lists and designated List 2 items as irrelevant by the F-instruction, there will be significant directed forgetting cost on List 2 items due to retrieval inhibition, whereas there will be no benefit on List 1 items.

Considering, that according to the reset-of-encoding hypothesis, the benefit in the standard directed forgetting procedure is due to the improved encoding of the post-instruction items, improvement in recall of the relevant information when these items were encoded before the to-be-forgotten-items is not expected. The context-change account (Sahakyan & Delaney, 2003), however, would hypothesize different results, as this account assumed that the F-instruction changed the internal context of the participants and this internal context-change caused forgetting all pre-instruction information. Consequently, based on the context-change account it was hypothesized that both List 1 and List 2 would be forgotten if the F-instruction followed the study of List 2 items.

It is detected that the size of DF cost was sensitive for the specific form of the F-instruction, and for the stimuli presented in the two study lists (see MacLeod, 1998; see also Sahakyan & Delaney, 2003). Therefore, first, we wanted to find significant directed forgetting cost and benefit with the specific design and procedure that were applied in our laboratory.

Then, using this procedure, we aimed to replicate the lack of DF cost of the F-instruction without consecutive second list learning. This could be important because the only published study in this regard was Pastötter and Bäuml (2007), as Gelfand and Bjork (1985) was a poster presentation. Then, in the third experiment using the same instructions and same stimuli then in Experiment 1 and 2, we aimed to find a significant cost of F-instruction without post-instruction encoding.

VIII.3.1. Experiment 1.

The aim of Experiment 1 was to detect significant cost and benefit of F-instruction in a list-method directed forgetting paradigm with two study lists.

Method Participants

Altogether 60 (34 men and 26 women) Hungarian undergraduate students (native Hungarian speakers) participated in the first experiment between the ages of 18 and 28 years.

They were recruited at the Budapest University of Technology and Economics, Hungary and received extra course credits for their participation. All participants gave written informed consent. In each of the three experiments, participants were randomly assigned into either a

‘forget’ or a ‘remember’ group. Based on the post-experiment debriefing, the data of three participants were excluded from the analysis, as they were informed of the goal of the experiment.

Stimuli and Design

Two lists (List A and List B) of Hungarian words with moderate recall probabilities were used as stimuli (12 items/list). Half of the participants in both groups (forget, remember) were first presented with the List A items, whereas the remaining participants were first presented with the List B words.

Procedure

All experiments (Experiment 1-3) were conducted under the same conditions: in the same lab with the same experimenter. Participants seated at 60 cm from a computer display.

Stimuli appeared in random order on the computer screen (2000 msec/word) with a 1000-msec delay between them.

Participants were informed that they would see a list of words on the computer screen and were asked to memorize as many words, as they could. Immediately after they were given the List 1 items, participants in the forget group were given a between-list forget-instruction (F-instruction). They were told that the previously studied words were presented by a mistake and they were asked to forget the List 1 items. Before the presentation of the List 2 items, they were instructed to concentrate on the upcoming list and were asked to memorize as

151

many List 2 words, as they could. Immediately after the presentation of the List 1 items, participants in the remember group were told that they had completed studying the first list and that they would receive a second list of words that had to be remembered as well (between-list remember-instruction [R-instruction]).

Immediately after studying the second list, participants in both experimental groups were exposed to an 8-minute arithmetic distractor task that was followed by a free recall test.

Participants were first asked to recall the List 1 items, and then, the List 2 words. Participants were directed to recall first list items right away in order to eliminate output interference (see Anderson, 2005).

Results and Discussion

A mixed-design ANOVA was conducted on recall rates with Group (Forget/Remember) as a between-subjects variable and List (List 1/List 2) as a within-subjects factor. During post hoc analyses, independent-samples t-tests were used for the between-subjects factor.

The ANOVA revealed the predicted Group x List interaction (F[1, 57] = 15.52, p < .001,

²p = .21). Participants in the forget group recalled fewer List 1 words (t[57] = -2.52, p < .05, r

= -.32) and more List 2 items (t[57] = 3.31, p < .01, r = .40) than participants in the remember group. In brief, a between-list F-instruction impaired memory for the List 1 items (the cost of the F-instruction) and improved memory for the List 2 words (the benefit of the F-instruction), in comparison with a baseline condition when a between-list R-instruction was given for the participants (see Figure 1A).

Significant cost and benefit of F-instruction were found in Experiment 1. This is an important starting point for Experiment 2 and 3, as based on the dominant theoretical account of the field, the lack of directed forgetting effects (cost and/or benefit) was hypothesized in the following experiments, and this experiment gave evidence that the lack of directed forgetting effect could not be due to the specific procedure or to any item features of the present study.

Figure 1. Comparison of Recall Rates between the Forget and the Remember Groups in Experiment 1 and 3

Notes. (A) Experiment 1: in a standard list-method directed forgetting task, a significant cost of a between-list F-instruction was seen for the first study list (fewer recalled List 1 words in the forget group than in the remember group) with a benefit of the F-instruction for the second study list (more recalled List 2 words in the forget group than in the remember group). (B) Experiment 3: when the F-instruction was given for the second study list after its presentation, a significant cost of the forget-instruction was found for the second list without a benefit of the F-instruction for the first list when participants in both groups were first asked to recall the List 2 items, and then, the List 1 words.

Error bars represent standard error of the mean (SEM).

VIII.3.2. Experiment 2.

The aim of Experiment 2 was to replicate the main results of Gelfand and Bjork (1985) and Pastötter and Bäuml (2007), as in these studies were found that there were no cost of F-instruction if the to-be-forgotten list was the only study list in the forget condition.

Method Participants

Altogether 60 (30 men and 30 women) Hungarian undergraduate students (native Hungarian speakers) participated in the second experiment between the ages of 18 and 28 years. They were recruited at the Budapest University of Technology and Economics, Hungary

153

and received extra course credits for their participation. All participants gave written informed consent. Based on the post-experiment debriefing, the data of three participants were excluded from the analysis, as they were informed of the goal of the experiment.

Procedure

We used the same procedure as in the first experiment until the point when participants were given a second study list in Experiment 1. In Experiment 2, after the presentation of the List 1 items, participants were given either an F- or an R-instruction, but there was no second list following the F-/R-instruction. Instead, participants in both groups were given the 8-minute arithmetic distractor task, and then, the free recall test.

Results and Discussion

In Experiment 2, recall rates were compared between the groups (forget vs.

remember) by conducting an independent-samples t-test. When we compared the forget group’s recall rate (M = 35.2%, SD = 15.7) to the remember group’s performance (M = 38.7%, SD = 18.6), we did not find any reliable difference between them (t[58] = -0.77, n.s.). These results support that an F-instruction after the presentation of a list without a consecutive study list is not enough to the suppression of the to-be-forgotten items.