Studies (Edexcel A-Level Psychology): Revision Notes
Schmolck et al. (2002) HM and Other Brain-Damaged Patients
Overview
Case studies examining brain-damaged patients have proven valuable in understanding the nature and function of human memory. Research into amnesia patients has helped establish which brain regions are responsible for storing and processing newly encoded information and recalling previously learned material. It is well-established that damage to the temporal lobe is associated with memory loss, particularly anterograde and retrograde amnesia, though often well-established long-term semantic knowledge remains intact.
Anterograde amnesia refers to the inability to form new memories after brain damage, while retrograde amnesia involves the loss of memories formed before the injury. Understanding these distinctions is crucial for interpreting the findings of this study.
Theorist, title, year
Heike Schmolck, Elizabeth Kensinger, Suzanne Corkin and Larry Squire (2002) - Semantic knowledge in patient HM and other patients with bilateral medial and lateral temporal lobe lesions.
Aim
The researchers sought to investigate the effects of specific brain damage on semantic memory by comparing case studies of brain-damaged patients with a control group of individuals without neurological impairment.
The study aimed to:
- Test the relationship between semantic test scores and temporal lobe damage
- Determine whether Henry Molaison (HM) was unique in how his brain damage affected his memory compared to similar damage in other cases
Participants
The study compared six participants with amnesia to eight control participants. The control group was matched for age (74 years old), sex (male) and education (12.4 years) to the amnesia patients, with the exception of one female participant.
The six amnesia patients were divided into groups based on the type of brain damage they had suffered:
- HF group (n=2): Brain damage largely restricted to the hippocampus, caused by encephalitis (inflammation of the brain causing damage to brain structures)
- MTL+ group (n=3): Suffered encephalitis resulting in large medial temporal lobe and anterolateral temporal cortex damage (middle, front and side regions)
- HM: Had medial temporal lobe damage with some lateral temporal lobe damage following surgery to resolve his epilepsy
All patients had suffered bilateral (both hemispheres) damage to varying degrees. This bilateral nature of the damage was crucial for understanding the full impact on semantic memory function.
A comprehensive biography of each patient was compiled to provide context for their performance on memory tasks.
Apparatus
Nine core tests were conducted over three to five different sessions with participants. Seven of these tests came from the Semantic Test Battery and two tests were constructed by the researchers themselves.
All tests were based on line drawings of 24 animals and 24 objects. The 48 line drawings were further categorised into groups of eight: domestic land animals, foreign land animals, water animals, birds, electrical household items, non-electrical household items, vehicles and musical instruments.
This systematic categorization allowed researchers to test semantic knowledge across different domains and identify whether certain types of objects (living vs non-living) were more affected by specific brain damage patterns.
Procedure
Tests
A total of 13 tasks were designed to measure semantic knowledge related to identifying, sorting or defining the line drawings. Four additional semantic tests were conducted on some patients and control participants (HM only received test 10 of these additional tests).
Tasks 1-4: Pointing to or naming a picture Participants were asked to point to or name a picture when given the name or a description of the object.
Task Example: Picture Naming
A participant might be shown a line drawing and asked "What is this?" or given the name "elephant" and asked to point to the correct picture among several options. This tests both recognition and retrieval of semantic knowledge.
Task 5: Semantic features Participants were asked to answer yes/no questions about the physical and associated features of an object.
Tasks 6 & 7: Category fluency and sorting Participants were asked to name or sort into categories as many examples within a category or class (living/non-living) of objects without a picture cue.
Tasks 8 and 9: Defining task Participants were given the name or picture of the 24 less common objects and asked to provide a definition.
Task 10: Pyramid and palm tree test Participants were given a target picture and two test pictures and asked which test picture went with the target picture. For example, a target picture of a saddle was presented with two test pictures of a horse and a goat, and participants were asked to say which test picture went with the target.
Task 11: Object/non-object discrimination task Participants were asked whether the object presented to them was real or not.
Task 12: Colouring object task Participants were asked to colour 28 line drawings of objects using appropriate colours from a selection of four coloured pencils.
Task 13: Nouns and verbs test Participants were given a fill-in-the-gaps exercise designed to test knowledge of regular and irregular verbs and tenses.
The percentage of correct responses was scored for all tests other than test 6, 8 and 9, which were recorded and transcribed and were given an accuracy rating between 0 and 4 by the researchers. These transcripts were also assessed for errors in grammar, expression, confusion and word intrusions.
Inter-rater reliability (the degree of agreement and consistency between raters about the thing being measured) was established for the scoring to ensure objectivity and consistency.
Findings
Tasks 1-9
Patients with damage restricted largely to the hippocampus (HF group) were able to name, point out and answer questions about objects they were given with considerable accuracy. They were also comparable to the control group when asked to generate examples of a given category or give definitions of objects.
Critical Finding: Hippocampus-Only Damage
The HF patients' normal performance on semantic tasks suggests that the hippocampus is NOT primarily responsible for storing semantic knowledge. This challenges earlier assumptions about hippocampal function in semantic memory.
Patients with both medial temporal lobe and anterolateral temporal cortex (MTL+ group) performed less well at naming, pointing out or answering questions about objects. They also had considerable difficulty generating examples in a given category. Notably, one MTL+ patient could not generate names of dog breeds despite previously being a dog breeder. They also had difficulty defining objects, often giving less detail which contained more errors.
HM's Performance Pattern
HM performed worst among these patients, and the MTL+ patients found it most difficult to identify and recall facts about living objects compared to non-living objects in all tasks. This living/non-living distinction would become important for understanding semantic memory organization.
When participants were ranked by their overall performance on these tasks, their rank appeared to correspond directly with the extent of their brain damage. Specifically, damage to the anterolateral temporal cortex seemed to cause impairment in semantic knowledge.
Other semantic tests (tests 10-13)
When asked to decide whether an object was real or not, two of the MTL+ participants performed well, but one made eight errors. All MTL+ patients scored well on the colouring task.
On the Pyramid and palm tree test, the MTL+ patients and HM scored either below the required 90 per cent accuracy or performed below the control group.
The MTL+ group were able to produce regular plurals and verbs but performed less well at producing irregular verbs and plurals. In contrast, HM performed well on both tasks, suggesting that the difficulty with irregular items is associated with anterolateral temporal cortex damage.
Conclusions
The MTL+ patient data demonstrates that damage to the anterolateral temporal cortex is consistent with a loss of semantic knowledge that results in a 'blurring' or overlap of conceptual knowledge that leads to confusion. This semantic knowledge is associated with the anterolateral region and is not associated with the medial temporal lobe.
Supporting Evidence from Semantic Dementia
This is consistent with patients who have semantic dementia (a degenerative neurological disorder resulting in loss of semantic memory), whose impairment is restricted to the anterolateral temporal cortex and the medial temporal lobe is relatively unaffected.
Additionally, MRI scans (brain scans that produce an image of the structure of the brain with detailed resolution) seem to suggest that the more progressed the disease, the greater the anterolateral damage.
HM as a special case
HM was similar to the MTL+ patients in tests of definitions, suggesting that his impairment had a similar physiological basis. However, in many respects his semantic knowledge was in the normal range in other tests.
HM's Unique Language Deficit
Unique to HM was the large number of grammatical errors he made during these tests. The researchers suggest that his deficit in language production was unlikely to be related to his temporal lobe damage but due to other factors during his childhood:
- HM suffered from a seizure at age 10
- He was from low socioeconomic status
- His schooling was interrupted
These factors could have contributed to poor language development.
The researchers conclude that the hippocampus is not involved in semantic knowledge because HF patients performed similarly to the control group. HM was less affected than the MTL+ patients, which leads to the conclusion that the anterolateral temporal cortex and not the medial temporal lobe is involved in semantic knowledge. The language impairment displayed by HM was unrelated to his neurological condition and probably due to his upbringing.
Evaluation
Strengths
Supporting evidence from multiple sources Findings from semantic dementia research, neuroimaging studies, brain stimulation experiments and unilateral lobectomy procedures all support the finding that the anterolateral and, in particular, the lateral temporal cortex is involved in semantic knowledge. This strongly reinforces the conclusion of this study.
Detailed case study approach
The use of multiple tests across different sessions with comprehensive biographical information allowed for thorough assessment of each participant's semantic memory abilities. This methodological rigour strengthens the validity of the findings.
Inter-rater reliability established The establishment of inter-rater reliability for scoring ensures consistency and objectivity in how the data was evaluated.
Weaknesses
Limited generalisability
Case studies of brain-damaged patients are rare and therefore small in number. The small sample size involved in this investigation limits the generalisability (the ability to apply findings to other people, situations and contexts) of the conclusions made.
Retrospective research limitations The special case of HM reported in this investigation was seen as a product of upbringing and events prior to his surgery for epilepsy. However, it could be argued that the individual differences found in this investigation demonstrate individual variation in neurology which may account for the differences between them.
Retrospective research, such as this study, cannot establish causal relationships between the injury sustained and the resulting impairments tested. The brain is adaptable and can compensate for injury. The findings of tests may reflect the ability of the brain to adapt to injury rather than the injury itself. However, prospective research is not possible as it would involve predicting those individuals who are likely to sustain such brain damage.
Ecological validity concerns The stimuli that are common to many cognitive investigations used to test semantic knowledge, such as the line drawings used, lack mundane realism and may not tap into semantic knowledge as it is used in everyday life. Such research may be said to lack ecological validity, as the findings cannot be generalised to everyday use of semantic memory.
Nature-nurture debate
Schmolck et al. believed that HM's language impairment had developed due to causes other than the neurological impairment caused by his surgery, that it was perhaps due to nurture rather than nature. It is possible that his low socioeconomic status and interrupted education had a negative effect on his language development. However, the nature-nurture debate will never be resolved in this case because HM's language was not tested prior to the surgery.
Wider issues and debates
Socially sensitive research
The use of case studies of brain-damaged patients can be regarded as socially sensitive research because it is sensitive for the individual patient and their family. The amnesia patients have suffered injury or disease resulting in fundamental impairment to possibly the most vital of human cognitive processes: memory.
Memory anchors us to our past and enables us to project ourselves into the future; it links us to those we are close to and enables us to function normally in everyday life.
Deficits in memory have catastrophic implications not only for the individual and their ability to live a full life, but also on their family responsible for their care. This is very sensitive when amnesia occurs later in life when families have to adapt to lifestyle changes and potential loss of the person they once knew.
Ethical issues
Doctor Scoville, the surgeon responsible for HM's surgery, was vilified for his reckless approach. However, the case of HM was fundamental to our understanding that memory is a distinct cognitive process, independent of language and thought.
HM's Legacy to Neuroscience
Damage to his hippocampus and temporal lobes enabled neuroscience to establish a location for memory in the brain. This is arguably the most important advance in our understanding of memory functioning and HM was the most researched individual in the field of neuroscience.
Remember!
Key Points to Remember:
-
Anterolateral temporal cortex damage causes semantic memory impairment – MTL+ patients with damage to this region showed consistent difficulties with semantic tasks, whilst HF patients with hippocampus-only damage performed normally.
-
The hippocampus is not involved in semantic knowledge – HF patients performed similarly to control participants, suggesting the hippocampus plays a limited role in semantic memory storage.
-
HM was a special case – His language difficulties were attributed to childhood factors (seizure at age 10, low socioeconomic status, interrupted schooling) rather than his neurological damage alone.
-
Performance correlated with extent of damage – When participants were ranked by performance, their ranking corresponded directly to the extent of anterolateral temporal cortex damage as shown by MRI scans.
-
Limitations include small sample size and ecological validity concerns – The rare nature of these cases limits generalisability, and the use of line drawings may not reflect real-world semantic memory use.