STUDY 1
A laboratory experiment investigating the effect of stimuli organization on
recall in short-term memory
Abstract
The aim of the current study was to investigate the effect of stimuli organization on recall in
short-term memory. A laboratory experiment with a repeated measures design was
conducted. 10 healthcare professionals were recruited as participants from a local nursing
home via opportunity sampling. The independent variable (IV) had two conditions. In
condition A, participants were presented with a list of 30 random words (IV1- no
organization). In condition B, 30 words organized into 5 categories were displayed (IV2-
organization). The dependent variable (DV) was the number of words accurately recalled.
Numerous research studies (Bower et al., 1969; Bower & Springston, 1970; Lange, 1973;
Miller 1956) reported that organization improved the functional capacity of short-term
memory and items were better recalled when clustered into meaningful units. Therefore, the
directional alternative hypothesis states that participants will be able to recall more words
accurately when they are presented in an organized, rather than in a randomized fashion.
Comparing the mean recall of the 2 conditions (condition A= 8.2/ condition B= 12) has
revealed a difference of 3.8 points, which shows that participants’ recall was aided by
information organization.
Introduction
Memory has been defined as ‘’the retention of learning and experience’’ (Gross 2010,
p.257). The idea that there are qualitatively different memory systems was first articulated
by William James, who introduced the notion of primary and secondary memory, short-term
(STM) and long-term memory (LTM) respectively (1890, as cited in Jonides et al. 2008). The
first prominent successor of this view was the multistore model of memory proposed by
Atkinson and Shiffrin (1968), which depicted both STM and LTM as unitary systems. Baddeley
and Hitch (1974) proposed the working memory model, which suggested that STM had
different components for different types of information. The central executive is its main
component, which deals with cognitive tasks and coordinates the operation of the two slave-
1
,systems. The phonological loop deals with speech-based information, whereas the
visuospatial sketchpad processes visual and spatial information. The episodic buffer is the
newly added component, which is believed to integrate information from the CE and LTM
(Baddeley 2000).
It has been demonstrated by measures of both digit span and free recall that STM has a
very limited capacity. Murdock (1962), who used measures of free recall to assess STM
capacity, reported that items at the beginning and at the end of a list were better
remembered than those in the middle. This phenomenon became known as the serial
position effect. Primacy effect supposedly occurs because items at the beginning are
transferred to LTM via rehearsal, from where they can be retrieved. Items at the end are
believed to be better recalled because they are presumably still held in the STM. The recency
effect estimates an STM capacity of around 2-3 items. However, Miller (1956, p.81) suggested
that STM can hold 7 (+/- 2) pieces of information at a time arguing ‘’this number assumes a
variety of disguises, …, but never changing so much as to be unrecognizable’’. Yu et al. (1985),
who investigated the STM capacity for English and Chinese language materials, also found
that memory span for chunks, which could be encoded acoustically, was around 6 or 7 in both
languages.
Miller suggested that STM capacity could be increased by ‘chunking’, that is organizing
information into larger, meaningful units by relating them to information stored in LTM. As
Bower (1970, p.32) argued ‘‘items grouped together come to be recalled together as an inter-
associated cluster’’. According to Miller, the memory span seems to determine the total
number of ‘chunks’ that can be held in STM, and not the actual amount of information that
can be clustered into these units (Mayzner & Gabriel 1963, p.163). Several research studies
(Bower et al., 1969; Bower & Springston, 1970; Lange, 1973) demonstrated that items were
better recalled when they were organized into categories or hierarchies. On the other hand,
Dempster (1981) found no conclusive evidence that grouping and chunking of information
would influence individual differences in memory span. The current study investigates
whether organizing items into categories improves recall in STM.
Based on previous research, the experimental hypothesis states that participants will be
able to accurately recall more words when they are organized into categories than when they
are presented in an unorganized fashion. The null hypothesis states that organization of
2
, words into categories will have no effect on the number of words accurately recalled. Any
difference identified is attributable to chance factors.
Method
Design
To ensure a high level of control over confounding variables and reduce individual differences,
a laboratory experiment with a repeated measures design was conducted. In condition A,
participants were presented with a list of 30 random words (IV1- no organization). In
condition B, participants viewed 30 words organized into 5 categories (IV2- organization). To
eliminate order effects, counterbalancing was used with 5 participants completing condition
A first followed by condition B, and the other 5 doing condition B first followed by condition
A. In both conditions, the DV was the number of words accurately recalled.
Participants
10 healthcare professionals (6 females and 4 males) were recruited as participants from a
local nursing home environment via opportunity sampling. Age range: 21-50+.
Materials
Materials used include: Briefing/ Consent Form (Appendix A), Word Presentations for
condition A (no organization) (Appendix B) and condition B (organization) (Appendix C),
Answer Sheet (Appendix D), Debrief Form (Appendix E), a timer, a computer with a projector
and pens.
Procedure
All participants were briefed prior to the experiment and were assured that the cognitive
tasks used were not intelligent tests and that their data would be kept confidential.
Participants were informed that they had the right to withdraw from the experiment at any
given time. Following the collection of informed consent, 5 participants were tested in
condition A first followed by condition B. After viewing the word presentation of the list of 30
random words for 35 seconds, they had 1.5 minutes to recall as many words as they could in
any order. Participants were tested in condition B in the exact same manner using the
organized word list. The other 5 participants completed condition B first followed by condition
3
