Animal Models of Remembering: Delayed Matching to Sample & Radial Arm Maze

Delayed Matching to Sample

Delayed matching to sample is a procedure used to study remembering in animal models. It builds upon the matching to sample procedure, which is employed to investigate categorization, concept formation, and the development of equivalence classes.

Procedure

The basic procedure involves:

Presentation of Stimulus: A stimulus is presented to the subject (e.g., a pigeon).
Removal of Stimulus: The stimulus is removed for a specific delay period.
Presentation of Stimuli: The original stimulus and one or more other stimuli are presented.
Identification: The subject must identify the original stimulus.

Example with Pigeons

A pigeon is presented with a red light on a center key.
The pigeon responds to the red light, which leads to the removal of the stimulus.
After a delay, the original red light and other colored lights are presented on different keys.
The pigeon is rewarded for pressing the red key (identity matching).

Pigeons learn this rule using trial and error, typically starting with a zero delay during training.

Delays

During training, after the sample light goes out and the pigeon pecks, comparison keys light up immediately (zero delay).
Memory is investigated by extending the period between the sample and the presentation of the comparison keys, creating different delays.

Trial Sequence

Trial Start: A sample stimulus (e.g., a colored light) is presented.
Observing Peck (Optional): The pigeon may peck at the sample stimulus.
Sample Selection: The pigeon pecks at the sample stimulus.
Delay: The keys go dark for a set delay period.
Comparisons Appear: The comparison stimuli appear.
Correct Choice: If the pigeon chooses the correct stimulus, it receives food.
Incorrect Choice: If the pigeon chooses the incorrect stimulus, it does not receive food.
Intertrial Interval: A period of darkness before the next trial begins.

Multiple trials with different samples occur within a single session.

Other Animals

The procedure can be adapted for rats using levers instead of keys. The location of the lever can serve as the sample stimulus.

Forgetting Functions

Delayed matching to sample allows researchers to obtain a forgetting function, which plots performance accuracy as a function of delay. These functions are also obtained with humans.

Forgetting: The longer the delay, the worse the performance.
Common Across Species: This relationship between delay and accuracy is observed in many species.
Analogy to physical distance: The further away something is in time, the harder it is to discriminate, similar to how physical distance affects visual discrimination.

Factors Influencing Forgetting Functions

Encoding Phase

Researchers can manipulate factors that might impact the encoding phase of memory.

Sample Observation Time: Varying the duration for which the animal observes the sample stimulus could affect encoding.

Potential Outcomes

Decreased Encoding: If encoding is decreased, the forgetting function curve might shift downward (lower overall performance).
Increased Encoding: If encoding is increased, the curve might shift upward (higher overall performance).

If the lines are parallel, it suggests encoding is influenced, but the rate of forgetting remains the same.

Changes affect overall accuracy, not the rate of forgetting.
Suggests effects on encoding phase, but not the retrieval phase.

Example Study

An actual study varied stimulus exposure duration and found that:

Increased stimulus exposure increased overall levels of accuracy.
This increase is often interpreted as improved attention or encoding.

Data

The original stimulus was presented for 1, 4, 8, or 14 seconds before the delay.
Longer stimulus durations led to better accuracy at zero delay.
The rate of forgetting was similar across all durations.
The longer the stimulus was presented, the better the performance, suggesting increased attention or encoding, rather than a change in the rate of forgetting.

Retrieval Phase

Changes in the rate of forgetting are often associated with the retrieval phase.

A change in the rate of forgetting can be observed if the delay increases, affecting the slope.
Example: Introducing a distraction during the delay can change the forgetting function.

Distraction

Turning on a house light during the delay can distract pigeons, leading to a faster rate of forgetting.

Some manipulations can affect both encoding and retrieval, leading to changes in overall accuracy and the rate of forgetting.
This type of change is interpreted as a decrement in the retrieval phase of memory.

Summary

Encoding: Influenced by exposure duration to the sample stimulus.
Retrieval: Influenced by introducing interference during the delay period.

Expected Graph Changes

Encoding Differences: Start with different accuracy levels, but lines are parallel (uniform change); overall accuracy is affected at all delays.
Retrieval Interference: Start with the same accuracy level, but one line becomes steeper (non-uniform change); the rate of forgetting is affected.

Practical Applications

Understanding whether a memory decrement is due to encoding or retrieval problems can have practical implications.

Example: A study by Geoff White examined drugs prescribed for epilepsy that had memory performance side effects.
The study found that the drug affected encoding but not the rate of forgetting.
This suggests that interventions should focus on improving encoding rather than retrieval.

Working vs. Reference Memory

Working Memory

Reflects elements of stimulus control that change across trials.
Information that needs to be remembered for a short period within a trial.
Example: In a delayed matching to sample task, remembering the color of the sample stimulus for that specific trial.

Reference Memory

Refers to long-term or never-changing rules.
Stable elements of stimulus control that do not change across trials.
Example: In a delayed matching to sample task, understanding the rule of identity matching (selecting the matching stimulus in the comparison phase).

Limitations of Delayed Matching to Sample

It's difficult to determine whether a performance decrement is due to forgetting the matching rule or forgetting the sample.

Radial Arm Maze

Description

Consists of a central hub with arms radiating out from it.
Used to examine the duration and capacity of memory function in rats.

Procedure

At the start of the trial, the arms of the maze are baited (food is placed at the end of the arms).
The rat is placed in the middle and allowed to make a predetermined number of arm visits.
To maximize reinforcement, the rat must remember where it has been and where it has to go.

Example

Rats are given four choices in an eight-arm maze.
After a delay, they are allowed to make four more choices and need to remember where they went before.

Performance

Rats can remember which arms they visited for up to four hours.
Performance decreases after four hours and reaches chance levels at 24 hours.

Working and Reference Memory in Radial Arm Maze

Modified Procedure

A subset of the arms are always baited, while the remaining arms are left empty.
The rat learns which arms to visit and which to avoid over multiple days.
The arms are not rebaited during a single trial.
At the beginning of the trial, four arms have food, and four do not (always the same arms each day).
The rat needs to collect food from the four baited arms.

Error Types

Working Memory Error: Repeating an arm within a trial (forgetting where it has already been).
Reference Memory Error: Visiting an arm that is never baited (forgetting the rules of the game).

Example Errors

Working memory error: Rat revisits arm 3 within the same trial.
Reference memory error: Rat goes down arm 8, which is never baited.

Example Study

Charlotte Kay used the radial arm maze to study the effects of MDMA (ecstasy) on working and reference memory.

Compared MDMA to scopolamine (known to affect memory).
Scopolamine increased working memory errors.
MDMA increased reference memory errors.

Interpretation

MDMA affects cognition by making it difficult to remember the rules of the game rather than what happened recently.

Conclusions

Animals are capable of remembering.
There are different types of remembering.
Similar underlying memory processes may exist between humans and non-humans.
Environmental events influence remembering.
Remembering can be viewed as an operant (an observable behavior influenced by its consequences).