Mitchell, K.J., Johnson, M.K., Raye, C.L., Mather, M. & D’Esposito, M. (2000). Aging and reflective processes of working memory: Binding and test load deficits. Psychology and Aging, 15, 527-541.

The purpose of this article was to explore the difficulty older adults had with working-memory tasks.

Age-related binding deficits were found in 2 experiments. Older adults had more difficulty than younger adults when tested on 2 features instead of 1, whether or not binding was present.

Binding – a combination of remembered events

Episodic memory – remembering particular details of an event and attributing it to a source.

Prior research has shown that older adults have more difficulty remembering detail than younger adults (Burke & Light, 1981; Kausler, 1994; Spencer & Raz, 1995, for reviews).

OLDER ADULTS: have more trouble remembering a variety of detail:

• Spatial information
• Voice
• Color
• Temporal cues
• Person
• Cognitive processes

However, older adults may remember an object as well as younger adults, but they might have trouble remembering which object was in which location.

Many aspects of this article were inspired by the research of Chalfonte and Johnson (1996). The researchers compared feature memory (colors, objects, and location) and feature bindings (remembering object + location and object + color combinations). They also proposed that poor encoding abilities demonstrated by older adults comes from deficits in reflective component processes that are critical for establishing associations between elements of experience.

Two experiments examined the following:

Experiment 1

• Memory for individual features
• Memory for combinations of features

Experiment 2

• Used a procedure similar to Experiment 1, but examined factors at both encoding and test.

Method

Experiment 1

Participants

• 24 undergraduates (mean age = 19.5 years) participated for class credit.
• 24 older adults (mean age = 74.1 years) also participated. They were screened for Alzheimer’s Disease, Parkinson’s Disease, and any other abnormal conditions that could affect performance.
• The general memory abilities of the younger and older adults were comparable, despite education differences.
• 21/24 older adults participated in the Wechsler Memory Scale-Revised (Wechsler, 1987) and the California Verbal Learning Test (Delis, Kramer, Kaplan, & Ober, 1987). Table 1 shows that the older adults scores were within established norms.

Design

• 2 (age: young, old) x 3 (condition: object only, location only, or combination) x 2 (item type: repeated vs. non-repeated objects) mixed design.

Variables of Interest

• Age
• Young and older adults’ ability to remember combined object + location information as compared with their memory for individual features.
• The type of to-be-remembered information was manipulated within subjects
• Extent to which similarity of the objects across trials might affect the performance of older adults

Materials and Procedure

• Stimulus was presented
• Computer-based instructions were presented
• Responses and response latencies were recorded
• Subjects began the experiment at their own pace. They had computer instructions.
• Verbal instructions were also given.
• Older adults needed more explanation.
• Subjects were told to remember pictures and common objects.
• At the start of the trial, the cue was presented in the middle of the screen for 500 ms.
• 3 x 3 grids contained drawings of different objects (e.g. bell, whistle, key: Snodgrass & Vanderwart, 1980).
• Each object appeared in one of 8 colors.
• The object was sized to fill the square.

Blocks of Trials – Participants were told:

• What objects were presented (object trails).
• Which locations were filled (location trials).
• Object + location together – were combination trials
• Blank

The screen emphasized that each trial was independent. The participants were tested on information that they were told to study. There were 2 practice trials that used abstract shapes as objects.

Blocks and Block Orders

• Total of 12 blocks of trials (4 object, 4 location, 4 combination).
• Each block contained 8 trials
• Same for each participant (object only, location only, combination, location, combination, object).
• There was a rest period after 6 blocks (2-3 minutes).
• The second set of 6 blocks followed in same order.
• Tasks became increasingly difficult.

All Trials

• 8 s after 3^rd study, participants were given recognition test on single object, location, or object + location combination.
• There were an equal number of targets and lures.
• Each test was cued by the word "TEST."

Targets

• Object trails: black and white line drawing of studied object in center of 3 x 3 grid.
• Location trials: filled black circle – little smaller than the objects – presented in one of the squares that had been filled for trial.
• Combination trials: black and white line drawing of object in its correct studied location.

Lures

• Object trails: black and white line drawing of studied object in center of 3 x 3 grid.
• Location trials: filled black circle presented in a location not filled on that trial.
• Combination trials: black and white line drawing of object studied on that trial located in a square that had been filled by a different object on that trial.

Test Probe

• Presented for 2 s
• Participants told to respond quickly, but accurately.
• Pressed Y for yes and N for no
• Participants could respond anytime test probe appeared
• Computer collected responses and response latencies (screen was blank)

Results and Discussion

Table 2 showed proportions of hits and false alarms and standard errors of the mean for experiments 1 and 2 by condition and age.

• No differences were found for between age groups in memory for objects or locations, for main effect of age, and for the interaction in Figure 2a.
• Older adults performed significantly below young adults in the combination condition, but not significantly below young adults in the single feature condition.
• Older adults poor performance in the combination condition happened because they are more likely than young adults to produce false alarms in response to re-paired lures (See Table 2).
• Older adults had no problem remembering features studied.
• Accurate rejection of lures requires memory for features and memory for how the features were paired during study.

Figure 3

• This figure shows the lure-target difference scores for combination and feature trials for each age group (with standard errors).

Old and Young – Feature-only condition

• Responded faster to lures than to targets (as shown by negative scores in Figure 3).

Old and Young – Combination condition

• Responded slower to lures and targets (as shown by positive scores).

Older Adults had Difficulty:

• Rejecting the combination lures.
• They had a deficit in processes that support binding and encoding.

Latency and Accuracy Data Show:

• Older adults have difficulty pulling object and location together in this working memory task.

Experiment 2

The same stimuli and trial structure that was used in Experiment 1 was used in Experiment 2. However, there were some differences. This experiment focused on remembering combinations of features – which is a more complex task. Experiment 2 evaluates the amount of information older adults can maintain (maintenance load hypothesis) and the difficulty with remembering multiple features (test load hypothesis). Conditions are summarized in Table 3.

Participants

• Sixteen undergraduates (mean age = 19.6 years) and 16 healthy older adults (mean age = 75.1 years) came from the subject pools used in Experiment 1.
• Completed education was equivalent.
• Scores from standard memory tests were available for 12/16 older adults (See Table 1).
• Older adults were paid for their participation.
• Younger adults received course credit.

Design

• Study-feature – test feature
• Study combinations – test combinations
• Study feature combinations – test combinations
• Study combinations – test two features (See Table 3 for more detailed information)

Materials and Procedure

• Stimuli and trial times were the same as Experiment 1.

What’s Different?

• Participants were first given a single block of 16 trials of one of the single- feature combinations.
• Block of 16 trials of the alternate single-feature condition
• Studied only one feature (object of location)
• Verbal instructions with illustrations followed each block
• All trials began with the word "STUDY."

Test probes

• Same as Experiment 1, but the object for memory used did not have a grid when it appeared in the center of the screen.
• Study-feature – test feature blocks were followed by 104 trials; participants were given combinations of encoding instructions.

Results and Discussion

• The contrast in d’ scores show that no significant differences were found between the 2 groups on their memory for objects and locations only, main effect of age, and for the interaction (See Table 2).

Older Adults Again Exhibit a Decrement in Their Memory for Combinations of Features

• Older adults had poor performance on the combination test condition, once again.
• Unlike Experiment 1, young adults also had poorer performance on the combination test condition as compared to the study feature-test feature condition (See Figure 4).
• Yet, young adults still did better than older adults.
• Figure 5 shows the difference scores for each age group for each condition. Again, young and older adults responded more quickly to lures than to targets in the study feature – test feature condition (as evidenced by negative scores). Young and older adults responded more slowly to lures than to targets in the study combination – test combination condition. The positive scores demonstrate this.

General Discussion

The present study focused upon whether or not older adults experienced binding deficits as a result of disrupted initial encoding processes that are used to form a memory. Both experiments showed that older adults had feature-binding deficits in a working memory task.

As the number of features to remember for a task increased in difficulty, older adults performed more poorly. In accordance with the results found in this study, the lacking ability of older adults to recognize combinations of features (i.e. an object and its location) are due to age-related source memory deficits. Reflective processes were interrupted and resulted in poorer complex memories.

The evidence from the study also suggested that test load deficits exist in older adults. This would make it difficult for older adults to process and evaluate multiple pieces of information (multiple features). Reflective processes and encoding abilities may overlap; thus hindering performance in working memory tasks.