Implicit Memory - Conceptual (7)

Conceptual Implicit Memory and the Item-Specific-Relational Distinction (Mulligan, 2006)

Explicit vs. Implicit Memory

• Dissociations: There are multiple dissociations observed between explicit and implicit memory, which include:

  • Amnesic patients

  • Normally aging older adults

  • Individuals with depression and schizophrenia

  • Effects of pharmacological interventions (e.g., benzodiazepines)

  • Neuroimaging studies indicate different brain regions are active during explicit vs. implicit memory testing

• Semantic Processing Effects: The efficiency of memory retrieval can be influenced by semantic processing and the overlap between study and test conditions.

Transfer Appropriate Processing (TAP)

• Performance Dependency: The performance on a memory test is heavily reliant on how similar the processing is during the test compared to what was enacted during the encoding phase.

  • If the processing overlaps appropriately, performance is high.

  • Conversely, if the processing is different or inappropriate for the test context, performance suffers.

• Explicit vs. Implicit Memory:

  • Explicit memory is characterized by reliance on semantic/conceptual processing. Enhancement in conceptual processing at encoding (e.g., depth of processing) leads to better performance on tasks like free recall and recognition.

  • Implicit memory, however, depends on perceptual processing, requiring perceptual engagement with stimuli to achieve effects such as priming in stem completion or perceptual identification (including masked stimulus identification).

Memory System Categorization (Roediger, 1990)

• Modes of Processing:

  • Declarative (Episodic): Involves explicit memory tasks including free recall and recognition.

  • Procedural (Priming): Involves implicit memory tasks such as word fragment completion and perceptual identification depending on data-driven processing.

  • Meaning-based (Conceptually-driven): Tied to general knowledge and free recall tasks.

Perceptual Interference Effect

• Overview of Experiment 1 (Blaxton, 1989):

  • Studies by Nairne (1988), and Hirshman & Mulligan (1991) indicate a perceptual interference effect: disrupting the encoding of a word can lead to subsequently improved explicit memory compared to uninterrupted study.

  • Experimental design included presenting a word (for 100ms) followed by a mask (for 2400ms), or presenting the word for 2500ms unmasked.

  • This effect diminishes if perception is completed before the mask (generally after about 200ms).

  • No impact was noted on conceptual implicit memory when tested in relation to explicit memory tests which displayed marked effects (for example, comparing category cued recall versus category exemplar production).

Problems for TAP

• Key Effects and Issues Identified:

  • Perceptual interference effect

  • Picture superiority effect (Weldon & Coyote, 1996)

  • Differences noted between category exemplar production and category-cued recall

  • Investigated word association versus associate-cued recall issues

  • The concreteness effect (Hamilton & Rajaram, 2001) revealed distinctions between general knowledge and question-cued recall

  • Distinctiveness effect (Smith & Hunt, 2000) observed between varying recall types

  • Within-category serial position effects critically addressed by Mulligan & Stone (1999, Experiment 5).

  • Generation effect (Slamecka & Grad, 1978) highlighted the differences between semantic (e.g., antonyms) and non-semantic generation rules regarding conceptual priming and explicit memory effects.

Item-Specific vs. Relational Information

• Definitions:

  • Item-specific Information: Features unique to an item, characteristics that uniquely define it, and sensory images it elicits.

  • Relational Information: Features shared among to-be-remembered items in context, indicating how well an item connects to its context or other items in a set.

  • Significance: Relational information assists in guiding self-directed (free) retrieval, whereas item-specific information plays a crucial role in differentiating items from non-studied items during retrieval.

• Category Exemplar Production vs. Cued Recall:

  • Category exemplar production does not require distinguishing a generated item from all other exemplars. In contrast, cued recall necessitates differentiating between studied and unstudied items.

• Explicit Memory Benefit: The enhancement in explicit memory is driven by item-specific processing and is mediated by perceptual processing rather than rehearsal.

  • The magnitude of this effect remains consistent regardless of total duration (e.g., 900ms versus 2500ms).

  • Associative or relational processing occurs after perceptual processing (identification).

  • Enhanced effects are more pronounced in recognition tasks than in recall tasks, where recognition notably relies on item-specific processing.

  • Implications include reduced category clustering at recall and impaired memory for order, affecting repeated recall tasks, with gains attributed to item-specific processing and losses attributed to relational processing when moving from test to retest.

Generation and Conceptual Priming

• Impact of Generation:

  • Generation leads to a steady explicit memory benefit across tasks (free recall, cued recall, recognition). Nonetheless, it tends to disrupt relational and order information.

  • Sometimes effects can reverse when order information becomes critical for the explicit task.

• Semantic vs. Perceptual Cues: Only generation from semantic cues enhances conceptual priming, whereas generation from perceptual cues yields no benefited effect.

Experiment Insights (Mulligan, 2002)

• Design:

  • Words were either read or generated with specific tasks, including letter transposition (e.g., rgass) or fragment generation (e.g., gra_s).

  • Comparison involved implicit category-exemplar production versus explicit category-cued recall, with instructions adjusted between subjects.

• Priming Observations:

  • Conditions yielded similar priming effects regardless of whether words were generated or read.

Levels of Processing (LOP) and Conceptual Priming

• Relationships:

  • LOP has an impact on explicit memory but does not affect perceptual implicit memory as strongly.

  • LOP does influence conceptual implicit memory aligned with TAP theory since deep processing tasks frequently encourage item-specific processing.

• Implications: A robust interpretation of the item-specific-relational account suggests that item-specific processing should not result in conceptual priming in category-exemplar production, which inherently relies on relational processing.

  • Considerations about stimuli suggests that they may have contained salient relational (category) information during encoding or the task might have been influenced by explicit retrieval factors.

Research by Mulligan et al. (1999)

• Focus: Investigated the interplay between item-specific processing (related to LOP) and relational processing (influenced by list organization) during category-exemplar production and category-cued recall tasks.

  • Hence, comparisons were made regarding encoding tasks that were either random or blocked within-subjects, while list organization varied between subjects.

Conclusions

• Key Distinctions:

  • Item-specific processing is sensitive to item-specific information but not relational information; conversely, relational processing is sensitive to both item-specific and relational information.

• Effects Noted:

  • Several effects, such as perceptual interference, picture versus word sensitivity, concreteness, difference judgment sensitivity, levels of processing, and list organization all contribute distinctively to task outcomes.

• Distinctiveness Insight: Distinctiveness is not strictly synonymous with item-specific information; it may arise from analyzing item-specific characteristics.

  • Types of Distinctiveness: "Prima facie distinctiveness" refers to items being unusual in isolation, while "secondary distinctiveness" occurs based on the surrounding context.