The mechanism of human memory recall is neither a parallel nor a sequential retrieval of previously learned events. Instead, it is a complex system that has elements of both sequential and parallel modalities, engaging all of the sensory faculties of the individual. On an everyday level, issues about memory and recall affect everyone. It has a bearing on ramifications from the trivial to matters of life and death. Thus, a particular student might worry about his or her ability to remember ‘memorized’ material, a person might worry about losing his or her mind, and, there are the more troubling issue of diseases affecting memory such as Alzheimer’s disease. According to Tulving, episodic memory represents only a small part of the much larger domain of memory (Tulving, 1992, p.1). Specifically, episodic memory is the process involved in remembering past events. This paper is a review of research findings on episodic memory with specific attention to episodic memory in adults and infants.
Episodic Memory in Adults
In society, it is quite common for people in their golden years or even well before that, to worry about losing their memory. There is scientific evidence to support this notion of degradation of memory with age. It is now well known in neurology that brain cells die off as one ages. Verhaeghen and Marcoen (1993, pp. 172-178) found that the decline associated with age in relation to the ability to perform episodic memory tasks involving deliberate recall appears to be largely a quantitative rather than a qualitative phenomenon. The ability of older adults to recall individual items in lists, or ideas in texts could be predicted based on the performance by younger adults on the same tasks. From their data in a sample of 48 younger and 45 older adults, they postulated a relationship between recall and age with a median correlation of r = .88. The same item characteristics could be used to predict probability of recall by younger or older adults.
Kliegl and Lindenberger (1993, pp. 617-637) tested a model for correct recall and intrusions in cued recall of word lists. Intrusions are defined as false responses that were correct in an earlier list. The model assumes three exclusive states for memory traces after encoding; 1) with a list tag-with information about list origin, 2) without list tags, and 3) missing. Across lists, a trace can lose its list tag or it’s content. For retrieval, an optimal strategy of response selection was assumed. Younger and older laboratory-trained mnemonists participated in two separate experiments in which recall of permutations of a single word list across a single set of cued was held constant with individually adjusted presentation times. They reported that younger adults were more apt to have correct recall, while older adults were more susceptible to intrusions. Age differences were restricted to model parameters estimating the probability of generation of list tags.
In another study, Denney and Lasen (1994, pp. 270-275) compared the ability of youngsters and adults to remember specific information and / or information related to a particular context. They investigated the ability of individuals not just to remember some given information, but also the ability to connect specific information related to a context. The study involved eighty adults in a bimodally stratified age range. The subjects were either between 18 and 30 years of age or 60 and 85 years. They were shown slides containing a word related to a specific information. Denney and Lasen concluded that although the elderly have memory problems, it is not with regard to remembering specific information. In a study of adults with a similar bimodal age distribution, (eighteen men, 18- to 26-years-olds and eighteen men, 60- to 79-year-olds), Jennings, Nebes, and Yovetich (1990, pp. 77-91) hypothesized that older volunteers allocate more attentional resources to memory maintenance than do younger volunteers. Allocation of a resource supporting memory maintenance was inferred from performance and cardiovascular measures. Individuals performed a serial memory task both as a ’single task’ and as a ‘dual task’ that added simple reaction time stimuli. Jennings et al. found that items presented early or later in the serial list created relatively low and high memory loads, respectively. The results of this task-oriented experiment suggested that older men allocated greater attention to memory maintenance, particularly during high-memory-load items and activities. The older men exhibited a slowing of dual-task reaction time and increased heart rate during high- versus low-memory-load items. Cardiac and vascular reactions further suggested that memory maintenance is supported by phasic autonomic adjustments, and that with age, more of this support is required for adequate maintenance of episodic memory.
The foregoing studies were focused on understanding memory in healthy adults. Since failing memory and mental diseases have been shown to have some association, some studies have also examined the use of memory in the diagnosis of primary stages of dementia. Herlitz, Hill, Fratiglioni, and Backman (1995, pp. M107-M113) reported that a study of the efficiency of cognitive tests in diagnosing and staging dementia proceeds with the aid of cognitive parameters evaluating episodic memory, while visuospatial assessments help stage dementia. This finding was held by the researchers to imply a faster degeneration of episodic memory than visuospatial capacities.
Episodic Memory in Infants
Bauer and Dow (1994, pp. 403-417) conducted a series of three experiments that tested whether 1- to 2-year-olds generalize their knowledge of events to new "instantiations," and postulated one possible mechanism by which generalization is accomplished. In their first experiment, 16 and 20 month-old children enacted six separate event sequences. One week later the same children were tested for delayed recall. At delayed testing, the props used to enact one-half of the events were replaced by novel, functionally equivalent props. Children in both age groups used the new props to enact the events, thereby demonstrating spontaneous generalization. Experiments 2 and 3 tested whether generalization is accomplished through forgetting of the specific details of the original event. At Session 1, 16- (Experiments 2 and 3) and 20-month-olds (Experiment 2) enacted four separate events. One week later the same children selected from an array of props those used to enact the events in Session 1. Among the objects from which selection was made were functionally equivalent props of the sort used to assess generalization in Experiment 1. Children in both age groups performed reliably on the recognition-memory task. Results found that 16- and 20-month-old children have at their disposal the capacity to productively generalize their knowledge of events and to form specific, episodic event memories.
Episodic memory is the process of recalling personally experienced past events. The efficiency of this process is adversely affected by age. In a sense, this may explain the level of emotional distress that the aged and their kin and all others feel at the onset of failing episodic memory. Because it relates to individuals and their family and friends in a very personal way, it tends to rob them of past shared experiences in a way that other memory failures do not.