Classic and Contemporary Research into Memory

Memory Theorists

There are three main sections of human memory: sensory, short-term, and long-term. We have a tremendous capacity for what we can store and remember. However, there are also faults in our memories. They can be altered or changed by things in our environment or our expectations via schemas.

Theorists and researchers have investigated and experimented on many aspects of human memory to learn more about how it works and what affects it. Many theorists and researchers have provided valuable knowledge in studying human memory, starting with Hermann Ebbinghaus (1850-1909). He was the first to research memory experimentally.

Even though this research was unprecedented in memory, carefully developed methods and measurements have stood the test of time. Since then, much research has been conducted in this field. Many have researched and experimented on various aspects of human memory, some studying its intricacies and some examining it more holistically.

Some of the most famous theories are Atkinson and Shiffrin (1968), who introduced the Multi-Store Model of Memory; Baddeley and Hitch (1974), who proposed the Working Memory Model; and Tulving (1967), who identified the different types of long-term memories.

Famous Psychology Memory Experiments

There are many classic and contemporary studies in Psychology that various memory theorists have conducted. Some of the most famous ones will be briefly mentioned here and then explained in greater detail in their respective sections. These famous psychology memory studies include:

1. Working memory model: The influence of acoustic and semantic similarity on long-term memory for word sequences, by Baddeley (1966b)

2. Semantic knowledge in patient H.M. and other patients with bilateral medial and lateral temporal lobe lesions, by Schmolck et al. (2002)

3. Reconstruction from memory in naturalistic environments, by Steyvers and Hemmer (2012)

4. Developmental pattern of digit span in the Spanish population, by Sebastián and Hernández-Gil (2012)

Memory Recall Experiments in Psychology

Let's look at one of the most influential memory recall experiments, Baddeley's 1966 experiment on long-term memory.

Baddeley and Hitch put forward their Working Memory Model in 1974, a significant contribution to Psychology and Memory research.

However, even earlier, in 1966, Baddeley conducted experiments for his research on the influence of acoustic and semantic similarity of word sequences on long-term memory. This was one of the main human memory research studies that formed the Working Memory Model.

Baddeley 1966b had three independent variables:

1. Semantically similar or dissimilar words.

2. Acoustically similar or dissimilar words.

3. Performance before the delay period and after the delay period

The dependent variable was the number of correct words recalled in order. The researchers divided 72 participants into four groups:

1. Acoustically similar words

2. Acoustically dissimilar words

3. Semantically similar words

4. Semantically dissimilar words

Fig. 1 - People are more likely to remember information that is semantically similar than semantically dissimilar information

The researchers tested all four groups' performance before AND after the delay period. First, the participants did the task 4 times. They had an interference task during a 15-minute break. Then, the researchers tested how many words they could recall in the correct order. Here are the results of the study:

• The semantically similar group performed better than the semantically dissimilar group.

• The acoustically similar group performed better than the acoustically dissimilar group. However, the difference was slight.

• The semantically similar group performed better than the acoustically similar group.

Memory Research Studies

After classic studies, let's examine some more recent studies into memory research.

Memory Research: Schmolck et al. (2002) — Semantic knowledge

Patient H.M. suffered from epileptic seizures. As a treatment, his bilateral medial temporal lobes, including the hippocampus, were removed. After the treatment, H.M. couldn't form new long-term memories, which led to the conclusion that the hippocampus was essential for memory formation.

Schmolck et al. (2002) wanted to find out about semantic knowledge in H.M. and patients with lesions or damage in the same location (of brain area). His research design included an Independent Variable (IV) of damaged/lesioned brain region (naturally occurring IV) and a Dependent Variable (DV) of scores on semantic memory tests.

There were six participants and eight control group members. Two participants in the experimental group had damage to the hippocampus, and three had large lesions in the medial temporal lobe and other lesions of the lateral temporal cortex. One participant (H.M.) had medial temporal lobe damage.

Fig. 2 - Brain scan images are used in clinical and research settings to identify the location and extent of brain damage.

Participants had to complete nine tests with line drawings of 24 animals and 24 objects and their names. First, these animals and objects were subcategorised, e.g. into six land animals. Then, the participants had four additional tests. These tests measured different aspects of the patients' semantic knowledge in depth.

Here are the results of the study:

• The hippocampus patients and the control group had similar performances.

• The three patients with medial temporal lobe (MTL) damage had mild to moderate impairment on most tests.

• H.M. performed better than the MTL+ group but not in other patients with only hippocampus damage.

• There was a correlation between the extent of lateral temporal cortex damage and test performance.

These results suggested that damage to the lateral temporal cortex causes impairments in semantic knowledge. Based on this, we can conclude that semantic and episodic memories are located in separate brain parts.

Memory Research: Sebastián and Hernández-Gil (2012)

Digit span tests have been used in memory research for a long time. The measurement got more specific when Baddeley and Hitch (1974) introduced the Working Memory Model.

Jacobs (1887) conducted one of the first studies of digit span.

Jacobs found that adults have an average digit span of 9.3 and an average letter span of 7.3.

The theory behind the digit spans was that children learn to process information faster as they grow up. Consequentially, the digit span should increase with age.

Sebastián and Hernández-Gil (2012) wanted to determine how the phonological loop develops as children grow older. They recruited 570 Spanish children between the ages of 5 and 17 for their study. The researchers administered the digit span test. The aim was to determine the children's ability to retain information in their short-term memory and compare it to British children.

Sebastián and Hernández-Gil found that as children grew older, their digit span increased. Also, their working memory continued to develop up to when they were 17 years old, compared to the age of 15 in British children.

Memory Research: Steyvers and Hemmer (2012)

The researchers wanted to investigate the interaction between episodic memories and prior knowledge (semantic memories), i.e. how prior knowledge could be used to reconstruct memories for images of everyday scenes.

Phase 1 involved establishing what a person expects in a scene based on prior knowledge. Participants had to name objects they'd expect to find in five naturalistic settings, e.g. an office. Then, they had to enter the items into a computer as a list for one minute per scene. Afterwards, a separate group observed 25 images of five scenes. They had to recall objects they saw.

Phase 2 involved 49 new participants who were shown ten of the images used in the previous phase (from those five scenes) for either two or ten seconds. This phase aimed to test prior/semantic and episodic knowledge, respectively. The study results revealed that:

• Incorrect recall of items that were expected to be in the scenes was 9%

• Incorrect recall of items not expected to be in the scene was 18%

• The average output was 7.75 items for the two-second condition

• The average output was 10.05 items for the ten-second condition

• Overall, when scenes represented what the participants expected, memory was above or around 80% accurate

Based on these results, semantic knowledge can help accurately recall episodic memories of naturalistic settings. This way, more cognitive resources are available for other tasks. The study itself ensured the results had high ecological validity (as it used naturalistic environments), and the researchers controlled other factors that could affect their experiment.

Memory Experiment Ideas in Psychology

Can you think of any memory experiment ideas? Below is a list of potential memory experiments.

• An experiment can be created to test the memory letter span of people whose first language is not one with the English alphabet. If your first language is Chinese and you learn English as a second language, you must learn an entirely new alphabet. Would people like this have a shorter letter memory span?

• Create an experiment studying memory recall on acoustically and semantically similar words but have the word length varied. Would longer, less well-known words have a lower recall?

• An experiment that studies the effects of the length of a recall delay.