10 terms

HLM: Transfer-Appropriate Processing


Terms in this set (...)

Transfer-Appropriate Processing (TAP)
+ Assumption
The situation that occurs when the initial processing of information is similar to the type of processing required by the subsequent measures of attention.

Assumes that memories are represented in the brain in terms of the cognitive operations engaged by an event as it is initially being processed and that successful memory retrieval occurs when those earlier operations are recapitulated

Roediger, Blaxton etc.
2 Main Implication of TAP
(1) Success of encoding task depends critically on how memory will be tested - it makes no sense to ask whether an encoding or retrieval task is good or bad

(2) Encoding is a 'by-product' of on-line processing therefore no single cognitive system is responsible for memory encoding -rather related brain activity reflect cognitive operations engaged by study task, not a stand-alone 'encoding circuit'
2 Types of TAP (explain)
(1) Data Driven- Processing of physical features of stimulus (many sub-varieties like visual/ auditory, object/word etc.) -AKA Perceptually driven

(2) Conceptually Driven- Processing of an items meaning
Blaxton (1989)
Used Jacoby's experimental design to show evidence that information precessing is distinct for perceptual and conceptually oriented stimuli.

Like in Jacoby, study phase consisted of 3 conditions: (1) No context (COLD); (2) A context (hot-COLD) or (3) generate (hot-?). After the study 5 P took part in test phases which consisted of 1 of 5 tasks:
(1) Free recall
(2) Semantic-cued recall- (cosmos-?)- UNIVERSE
(3) Graphemic-cued recall- (unversed-?) -UNIVERSE
(4) Word-fragment completion
(5) General knowledge test- What was the Big Bang said to have created? UNIVERSE

1-3 are EXPLICIT tests of memory
4-5 are IMPLICIT tests of memory

RATIONALE: If explicit tests behave one ways and implicit another way, results within in category should be consistent. Namely, Explicit tests should be most beneficial for generate condition, then context then no context while implicit tests should be most beneficial for no context, then context and generate conditions. BUT THIS IS NOT WHAT HAPPENED

FINDINGS: Free-Recall, Gen Knowledge and Semantic-Cued recall all benefited from G-C-NC while Fragment Completion & Graphemic-Cued recall all benefited from NC-C-G. **This cuts across the implict/explicit memory distinction, suggesting processing involved isn't based on the nature of the memory (implicit vs. explicit) but on whether the memory is perceptually or conceptually based (AS TAP WOULD SUGGEST)

IMPLICATION: Former group is reliant on ceceptual processing, while latter relies on perceptual processing, so each group benefits from the same type of processing at study: generate= conceptual and NC=perceptual.

Further TAP evidence from a follow-up study (1992) found that the modality change (to auditory) only effected the perceptually-processed tests.
Johnson & Rugg (2007)
Neural evidence for TAP

TEST: Presnt P with object-word pairs to remember. Words appeared either in solution or against a landscape background. In the isolated word condition, P was told to mentally construct a sentence for the word. In the landscape condition, P was told to form an image of the word as an object somewhere in the scene. Afterwards, Ps given a word recognition test with words presented alone.

FINDINGS: At encoding very different patterns of activation occurred: Superior & lateral occipital cortex and fusiform gyrus were activated in the picture condition while ventromedial frontal cortex was activated for the sentence condition. At test, words perviously studied in picture condition evoked the same neural activity from brain and visa versa for isolation condition

IMPLICATION: Repetition of the brain activity used at encoding is involved with successful recall > neural evidence for TAP
Problems for TAP theorists (3)
(1) Retrieval Practice -Glover

(2) Generation & Revelation Effect- Mulligan & Lozito

(3) Aterograde Amnesia- Biggest problem
How is retrieval practice a problematic phenomenon for TAP theorists?
Some evidence suggests that final recall benefits most when practice is hardest.

Glover (1989)

TEST: O reads prose on fictitious country Mala for 10 minutes. Prose contains 24 idea-units. Alls Ps take memory test 5 days later that is either free-recall or recognition. P1 takes practice test @ day 3 thats either free-recall or recognition.

FINDING: Final performance is higher for mismatched condition (i.e. free recall @ day 3 and recognition @ day 5) which defies TAP since the encoding/ retrieval tests don't match
Why is anterograde amnesia problematic for TAP theorists?
Abundant evidence suggests that AA impaired on explicit tests, like recall and recognition but not in implicit ones. TAP has to account for this phenomenon as a difficult in CONCEPTUALLY driven processing (which is what recall and recognition test require according to Blaxton (1989)). AA, according to TAP theorists, are unable to benefit from conceptual transfer between study and test conditions -this is not the case though.

Vaidya et al. (1995)
Vaidya et al. (1995)
AA evidence that suggests that it is the instructions that are critical for accurate recall rather than the type of processing involved.

TEST: Vaidya et al (1995) construct lists of highly-associated word pairs (table-chair) and in the study phase presented AA and Cs with the target words (chair). In the test phase, Ps performed either an explicit word-associated cued recall test in which they were given the cue (table) and instructed to recall a strongly related study word) or an implicit word association tsp in which they were again given the clue but asked to say the first word that came to mind.

RATIONALE: Both tests require predominately conceptually-driven processing. Verified this by showing that the modality of presentation had not effect on test performance (as it did in Blaxton (1992)). Since these tests both require conceptual processing,TAP theory predicts that AA should be impaired on both.

FINDINGS: Normal priming in the implicit test combined with significantly poorer performance on the explicit test for AA.

Strong evidence that something's wrong with TAP
Modified TAP framework
The TAP theory has been able to account parsimoniously for an impressive range of data. However, the idea of classifying tasks purely along the data-driven/conceptually-drive dimension has probably now outlived its usefulness. This simple distinction does not seem to be fully adequate, but perhaps that should not be surprising. We will surely need richer ways of classifying memory tasks.
A match between encoding a retrieval operations (TAP) is generally beneficial for memory. In addition, however, there needs to be a role for specific encoding and retrieval factors. It does seem to be the case that different encoding operations can lead to generally better-elaborated memory traces. For example, deep (semantic encoding) is generally beneficial for recall. Similarly, different retrieval processes (such as greater retrieval effort) can lead generally to better recall. Finally, the amnesia results suggest that there is a role for the hippocampus and medial temporal lobes over and above the TAP principle. As we will see in the next lecture, the MTL seems to play a key role in binding events to their contexts and this binding is required regardless of the specific mental operations engaged. Without this binding, recall will not be possible even when encoding operations are recapitulated at retrieval.