Contemporary Study- Schmolck et al 2002

• This study investigates those with amnesia to determine which parts of the brain are mostly responsible for semantic memories. 
• Previous to this research, it was thought that the medial temporal lobe plays a part in semantic memories and the hippocampus is more responsible for episodic memories

• Anterograde Amnesia- Loss of the ability to create new memories AFTER the event that caused amnesia
• Retrograde Amnesia- Loss of memory BEFORE the event that caused amnesia (can form new memories)
• Encephalitis- inflammation of the brain causing damage to the structures of the brain 
• Bilateral- both hemispheres of the brain are involved
• Anterolateral- to the front and side of
• Lateral- towards the side of
• Medial- situated in the middle 

Aim:

• To investigate the effects of brain damage on semantic memory, comparing patients with brain damage and a control group
• To test the relationship between semantic test scores and temporal lobe damage
• To determine if HM was unique in the way his brain damage affected his memory differently compared to others with similar damage 

Method- Participants:

• 6 with amnesia with various brain damage (x2 mainly in hippocampus, x3 who suffered encephalitis and HM)
• 8 control participants who had 'normal' functioning brains, NO brain damage 

Participants were given 13 tasks on 3-5 separate occasions. All participants did tasks 1-9.

• Tasks 1-4: Pointing to or naming a picture
  -Had to name 48 items and identify the picture that matched a category
  -Scored and a % of correct answers found 
• Task 5: Semantic features
  -Asked 8 'yes' or 'no' questions about 24 items
  -Scored and a % of correct answers found 
• Tasks 6&7: Category fluency and sorting
  -Pictures into categories
  -Scored and a % of correct answers found 
• Tasks 8&9: Defining task
  -Had to describe pictures of the least common animals and objects 
  -Quality score- recorded and subjective analysis- 13 psychologists had to agree 

• Task 10: Pyramid and palm tree test
  -Scored and a % of correct answers found
• Task 11: Object/ non-object discrimination
  -Association of things
  -Scored and a % of correct answers found
• Task 12: Colouring object task
  -Identify colours
  -Scored and a % of correct answers found
• Task 13: Nouns and verbs
  -Word exercises
  -Scored and a % of correct answers found

Semantic Features:

• Control group- 91.9% correct
• Amnesic- 96.9% correct
• MTL- 80.9% correct
• HM- performed similarly to the poorest control- 85.4% but a normal score on non-living things- 95.8%

-Damage to the MTL and ATC perfomed less well at naming, pointing out and answering questions about objects

-HM --> performed the worst on defining objects, performed better than MTL but less well than HF patients and controls

Ranks:

• Control --> 6th
• Hippocampus --> 5th
• MTL --> 12th/13th
• HM --> 11th

• The deficits in semantic knowledge are most likely to be related to anterolateral temporal damage (semantic impairment most evident)
• Patients with hippocampus damage were comparable with the control group on other tasks unlike medial temporal and anterolateral temporal damage 
• When participants were ranked on performance of tasks, there was a direct correspondence between this and their level of brain damage 

The study provides evidence of reliability and shows HM does have a problem with grammar. It could be argued that it was due to his low socioeconomic background as well as his brain damage OR because he did not attend school as often as he should have because of his severe seizures.

Generalisability:

• The sample size is small. Only 6 participants took part, and only 1 female. This means it cannot be generalised to the wider population.
• The sample is narrow and limited, certain types of brain damage only. The areas affected were concentrated around the hippocampus, amygdala and they were all caused by viruses. This means the sample is limited and therefore it cannot be generalised to other brain damaged patients. 

Reliability:

• We can replicate the study because brain scans were used and each test was standardised. This means we can test for reliability.
• The research has inter-rater reliability. The researchers had to agree on some tests to ensure the measure of quality of description, so it is objective. The study consists of agreement so there is high reliability of the test scores.

Application to Real Life:

• It has real life applications as it provides professionals with more understanding of the brain. HM is fundamental to our understanding that memory is a distinct process. His damage enables neuroscience to establish location for memory and advance understanding.

Validity:

• It is highly scientific - objective ways of measuring damage. PET scans were used and %'s of correct answers. This shows clear cause and effect and reduces bias.
• The lab setting provides high levels of control. They all did the same tasks and matched pairs was used (educational background and age). This increased the chances of inferring cause and effect. 

Ethics:

• It could have caused participants distress because if they performed badly on a test they could have lacked confidence and been put down by it.