Distinction between episodic and semantic memory

Question 1

First suggested distinction

Answer 1

Tulving 1972

Question 2

Different levels of processing (physical, rhyme or meaning) at encoding improved episodic memory, but had no effect on semantic memory.

Answer 2

Jacoby & Dallas 1981

Question 3

Reading a word out of context at encoding improved semantic memory Generating words themselves improved episodic memory. (Study actually talks about data-driven vs conceptually driven processing)

Answer 3

(Jacoby &; Dallas 1983).

Question 4

In-depth assessment of HM’s ability to learn new semantic knowledge, testing his knowledge of famous people (e.g. Borg) and words (e.g. Jacuzzi) that became prominent before and after onset of his amnesia in 1953 On all tasks, HM showed good memory for knowledge acquired prior to the 1950s, but a severe deficit for knowledge encountered subsequently: suggests that HM’s anterograde amnesia affected both episodic and semantic memory

Answer 4

Gabrielli et al 1988

Question 5

Reported patient PZ, a university professor who became an amnesiac son after writing his autobiography Ability to recall events from his past life showed a clear temporal gradient, with recent memories very impaired but remote memories relatively spared. When knowledge of scientific terms was assessed, showed a similar retrograde amnesia, only able to define those acquired early in his career. NB potential confounding factors: scientific terms acquired early in career may have seemed much more meaningful/ been given more importance, as patient was still in formal education, studying for exams, just starting out in career etc.

Answer 5

Butters &; Cermak 1986

Question 6

Example of patients with brain damage that selectively affects the hippocampus showing selective anterograde and retrograde episodic impairments, sparing semantic memory.

• Reported children with selective hippocampal damage acquired early in life. (brief anoxia selectively affects the hippocampus, e.g. that caused by lack of O2 for a few seconds during birth)
• All children showed very impaired episodic memory (e.g. on drawing test), but performed well on tests assessing semantic memory (e.g. vocab)

Answer 6

Varga-Khandem et al 1997

Question 7

Example of patients with brain damage that selectively affects the hippocampus (sparing other temporal lobe regions) who can show selective anterograde and retrograde episodic impairments, sparing semantic memory:

Reported 2 patients:

• SS, with extensive medial temporal lobe damage; failed to acquire any new semantic knowledge however tested,
• PS: selective hippocampal damage. Could recognize new vocab and famous people that came into the public domain after her lesion.

Significant because replicate the findings of Vargha-Khamdem et al but in adults- otherwise children’s performance could be due to neuroplasticity.

Answer 7

Verfaellie et al 2000

Question 8

Evidence for seperate systems from patients with dementia.

Semantic task: shown series of photos, and asked to name them

Episodic: shown series of photos, imposed a delay, then asked which of these have you seen before? 2 different conditions:

Perceptually identical: just have to remember having seen the object. Perceptual representation is the same in the study and test phases, e.g. the same picture of a phone used.
Perceptually different: have to use semantic information to identify the object with the same function as the one just seen. Perceptual representation alone therefore cannot be relied on.

Results:

SD:

• Significant deficit in semantic task
• Ok in perceptually identical task (just episodic)
• Impaired in perceptually different task (where had to use semantic info as well)

Therefore: just as impaired at episodic memory as AD when the perceptual representation could not be relied on, and needed semantic memory as well.

AD:

• Impaired in both versions of episodic task
• Similar to controls on semantic task

Answer 8

Graham et al 2000

Question 9

Asked patients to perform a semantic task that involved identifying drawings associated with each other (e.g. a pyramid is associated with a palm tree rather than a fir tree). Patients’ episodic memory was later tested for the same drawings

Again found impaired episodic and preserved semantic memory in AD, and impaired semantic and preserved episodic memory in SD.

Answer 9

Simons et al 2002

Question 10

Concluded that episodic and semantic memory are sepearate, proposed a process-specific relationship between memory systems, the SPI model:

• (S): serial encoding
• (P): parallel storage
• (I): Independent retreival.

Image <-> Perceptual representation <-> Semantic <-> Episodic

Made a number of predictions:

• Preserved semantic memory should be possible in the absence of a functioning episodic memory system- as demonstrated in the selective hippocampal patients who are amnesiac but can learn new semantic knowledge (Varda-Khandem 1997, Verfaellie 2000)
• One should not be able to find patients who exhibit preserved episodic memory if their semantic knowledge is impaired: i.e. a double-dissociation between episodic and semantic memory should not be possible
• Problem: double-dissociation demonstrated in e.g Graham et al. 2000, Simons et al. 2002. If patients with Semantic Dementia have dysfunctional semantic knowledge, what info are they using to perform the episodic memory task?

Question 11

Proposed that perceptual info might be able to feed directly into episodic memory: Multiple Input Model.

Image <-> perceptual representation <-> semantic <-> episodic, PLUS perceptual representation <-> episodic

• Tested by comparing episodic memory in patients with SD for perceptually identical and perceptually different pictures (Graham et al 2000)
• Results are consistent
• Argue that episodic memory typically relies on multiple inputs from perceptual and semantic systems, and in the absence of meaningful semantic input, perceptual info alone can be sufficient to support successful episodic memory.