Overview

1. A poetic process theory
2. Principles of instance theory
3. Implementing the principles computationally
4. Examining the model

Jamieson et al. (2022)

Intuition pump

How does some cognitive ability work?

What is instance theory?

1. A collection of domain-general processing assumptions about how memory functions
2. Can be formalized computationally
3. Provides a theoretical basis to evaluate potential functional abilities (and limitations) of episodic memory systems
4. Can be explanatory and exploratory

Mnemic Psychology

Historical Sidenote

• Modern instance theory is broadly consistent with Richard Semon’s theory of memory developed in Mnemic Psychology (1923)
• Semon was a German theorist whose ideas were largely forgotten until they were translated to English (mid-century)
• See Schacter, D. L., Eich, J. E., & Tulving, E. (1978). Richard Semon’s theory of memory. Journal of Verbal Learning and Verbal Behavior, 17(6), 721-743.
• Semon’s ideas inspired Hintzman’s MINERVA 2

Poetic Psychology

Engram-complex

Physical Resonance

Memory as experiential resonance

Basic instance theory assumptions

1. People encode the details of individual experiences
2. Retrieval is similarity driven.

• the pattern in the present moment retrieves experiences with similar patterns from the past

Instance theory primers

• A collection of related process theories of memory and cognition
• Different researchers have different versions (broadly similar)
• Instance theories range in specificity from verbal theories to computational models

Example verbal instance theories

• Jacoby, L. L., & Brooks, L. R. (1984). Nonanalytic cognition: Memory, perception, and concept learning. The Psychology of Learning and Motivation, 18, 1–47.

• Kolers, P. A., & Roediger, H. L. (1984). Procedures of mind. Journal of Verbal Learning and Verbal Behavior, 23(4), 425–449.

Big picture idea

1. Lots of cognitive abilities may be understood in terms of processes involved in cued-recall.

Minerva 2

MINERVA 2 is a global-matching model of memory (Hintzman, 1984,86,88) that can be formally evaluated as a computer algorithm

• applied across numerous domains

• Frequency judgments (Hintzman, 1988)

• False Memory (Arndt, 1998)

• Selective memory deficits (Curtis & Jamieson, 2019)

• Age-related memory decline (DRYAD, Benjamin, 2012)

• Prototype abstraction (Hintzman, 1986)

Minerva 2 continued

• Artificial Grammar Learning (Jamieson & Mewhort, 2009)
• Implicit Sequence Learning (Jamieson & Mewhort, 2009b)
• Judgments of Likelihoods (Dougherty, 1999)
• Eyewitness identification (Clark, 2003)
• lexical access (Goldinger, 1998)
• Associative Learning (MINERVA-AL, Jamieson et al., 2010)
• Semantic memory (ITS, Jamieson et al, 2018)

Hintzman (1986)

MINERVA in Excel

Let’s look at how the model works