Dr Tom Mercer School of Applied Sciences

Tom Mercer is a Lecturer in Psychology. He joined the University of Wolverhampton in 2011 after completing a PhD at the University of Leeds.

Tom's research uses the experimental method to explore how new memories are created and maintained, with a particular interest in forgetting.

Why do we forget information over short periods of time?

Introduction

The experience of forgetting is very common and memories may be lost over very short periods of time (sometimes within just a few seconds). Forgetting can also have serious implications. For example, there are plentiful cases of convictions arising from mistaken eyewitness identification (e.g. Pezdek, 2012), which indicates a failure of memory. Understanding the sources of forgetting is therefore of central importance to psychology and has been since the nineteenth century (e.g. Ebbinghaus, 1885), yet debate concerning forgetting mechanisms continues. Following decades of research, there is now some consensus amongst cognitive psychologists that memory loss often arises through a process termed retroactive interference. Such interference occurs when the presentation of new information distorts or eradicates an existing representation and it is recognised as one of the most potent causes of forgetting (Nairne & Pandeirada, 2008). As such, it is surprising that the underlying process through which retroactive interference operates remains unclear. Several possible mechanisms have been postulated, but it is commonly assumed that interference is determined by the similarity between the memory representation and the disruptive, interfering material. The argument is that more similar interfering events cause a greater degree of forgetting. However, Wixted (2004) has argued that any distracting event can be a source of interference. There is some evidence for such “nonspecific” retroactive interference (e.g. Dewar, Cowan, & Della Sala, 2007), but it remains mysterious and in need of further investigation. 

Aims and objectives

This ERAS project aimed to explore the way in which retroactive interference works. More specifically, this research intended to answer the following questions:

1. Can any distracting event lead to interference and forgetting? If so, is this due to the formation of other, unrelated memories, or the capture of attention?
2. Is there any way to reduce or remove the detrimental effect of retroactive interference?

The studies designed for this project aimed to test core predictions of Wixted’s (2004) consolidation account of memory and forgetting, and examine the impact of nonspecific retroactive interference.

Methods

An experimental approach was adopted and all studies used the delayed recall paradigm. Participants were given stimuli to memorise (word pairs) and then completed an immediate recall test. During this recall test, individuals were given the first word from each pair and had to provide the associated second word. The intention was to examine how much they could remember directly after they had viewed all of the stimuli. After an interval, participants were asked to complete a second, unexpected recall test. Comparing performance over the two recall tests then allowed the amount of forgetting to be quantified.

To study the effects of retroactive interference, different types of distractor activity were introduced into the interval separating the two recall tests. The contents of this distractor task were varied and its effect on forgetting monitored. A “no-interference” comparison was established by using a control condition that removed the distractor task, leaving the interval unfilled (and without external sources of interference). The time at which the distractor task was presented also differed.

Research Design

In the first experiment, participants took part in a foreign language learning activity and were shown a list of Icelandic-English word pairs (Icelandic was chosen due to its unfamiliarity with the English-speaking volunteers). Two recall tests occurred, separated by a 20 minute interval. Half of the participants had to learn Norwegian-English word pairs during the interval (introducing similar interfering materials) whereas the other half were exposed to face pairs (introducing dissimilar interfering materials). These face pairs presented images of two people side by side (of the same race and sex). Participants had to try and remember which faces belonged together. Crucially, these distractor tasks were presented either directly after the first recall test (causing immediate interference), or after an 8 minute interval of wakeful rest (which delayed interference).

The second experiment also involved the learning of word pairs followed by two recall tests. Importantly, events occurring between these recall tests varied. In the control condition no distractor task was presented, in an attempt to remove interference. In the three experimental conditions, different types of distractor activity were introduced. One distractor task involved a verbal memory task (designed to contain material that was highly similar to the original word pair stimuli), whereas another distractor task used a nonverbal memory task (designed to contain material that was dissimilar to the original word pair stimuli). A third distractor task involved a very simple judgement task. Participants were shown images of faces and simply determined whether they depicted male or female individuals. This particular distractor activity did not place any demands on memory and was not cognitively demanding.

Summary findings

• Experiment 1 showed that the formation of new memories caused interference and forgetting, but the content of these new memories was unimportant. There was no significant difference in the amount of forgetting according to whether participants were exposed to similar (Norwegian-English word pairs) or dissimilar (faces) interfering material. This fits with the notion of nonspecific retroactive interference and supports the idea that even dissimilar distracting events can produce forgetting.
• Experiment 1 also suggested that delaying interference alleviates forgetting. Individuals exposed to interference immediately after the first recall test forgot 15.57% of word pairs, but this dropped to 6.83% when interference was delayed. An 8 minute period of wakeful rest was capable of protecting new memories from subsequent interference.
• Experiment 2 found that even the passage of time can lead to significant forgetting, since there was a notable decline in recall performance in the control condition (which did not contain a distractor task). This finding was unexpected since time itself is not typically predicted to lead to forgetting.
• Subsequent pilot work indicated that mind wandering and daydreaming may actually be responsible for such time-based forgetting. In an unstimulating environment (like the control condition), participants may allow their mind to wander onto unrelated thoughts, which may act as a form of “internal” retroactive interference, making it difficult to retrieve specific memories.   

References

Dewar, M. T., Cowan, N., & Della Salla, S. (2007). Forgetting due to retroactive interference: A fusion of Muller and Pilzecker’s (1900) early insights into everyday forgetting and recent research on anterograde amnesia. Cortex, 43, 616-634.

Ebbinghaus, H. (1885). Uber das Gedächtnis. Leipzig: Dunker & Humbolt.

Nairne, J. S., & Pandeirada, J. N. S. (2008). Forgetting. In H. L. Roediger, III. (Ed.), Cognitive psychology of memory. Vol. 2. Learning and memory: A comprehensive reference (pp. 179-194). Oxford: Elsevier.

Pezdek, K. (2012). Fallible eyewitness memory and identification. In B. L. Cutler (Ed), Conviction of the innocent: Lessons from psychological research (pp. 105-124). Washington, DC, US: American Psychological Association.

Wixted, J. T. (2004). The psychology and neuropsychology of forgetting. Annual Review of Psychology, 55, 235-269.

Post ERAS Success

Following on from the ERAS Project Tom has a new publication:

Mercer, T. (2014). Wakeful rest alleviates interference-based forgetting. Memory. Advance online publication. doi:10.1080/09658211.2013.872279