Mikey Does

Michael Ullman

Definition:

Michael Ullman is Professor of Neuroscience at Georgetown University, known for developing the Declarative/Procedural (DP) model of language — a neurocognitive theory arguing that language learning and processing rely on two biologically distinct memory systems: the declarative memory system (supporting the mental lexicon and factual knowledge) and the procedural memory system (supporting grammatical rules and sequential processing). The DP model makes specific, testable predictions about how L1 and L2 speakers use their brains differently and about what clinical populations reveal about the architecture of language.

In-Depth Explanation

The Declarative/Procedural model distinguishes two long-term memory systems with different neural substrates, developmental trajectories, and functional profiles:

Declarative memory is associated with the hippocampus and medial temporal lobe. It stores facts, episodes, and learned associations — including the mental lexicon: words, their phonological forms, their meanings, and irregular forms (went, mice) that must be memorized individually because they cannot be derived from rules.

Procedural memory is associated with the basal ganglia and frontal cortex. It supports rule-based sequential processing — the kind of implicit, automatic computation required for inflecting regular verbs (walked, walks) and building syntactic structures. In native speakers, procedural memory handles the “calculate-on-the-fly” aspects of grammar: given a stored root, apply the rules to generate a well-formed output.

In L1 acquisition, Ullman proposes a developmental pattern: early vocabulary and irregular forms are learned declaratively (memorized items), while grammatical rules become proceduralized with increasing linguistic experience. The mature L1 speaker uses declarative memory for lexical retrieval and procedural memory for morphosyntactic computation — a division of labor that produces the fast, automatic, effortlessly accurate grammatical processing characteristic of native speakers.

In L2 acquisition, the picture is different. Adult L2 learners — who come to the language without the procedural entrenchment of a native speaker — show greater reliance on declarative memory even for grammatical rules. A learner who knows that Japanese marks past tense with -た (-ta) may initially handle this as an explicit declarative fact to be retrieved rather than a proceduralized rule applied automatically. With sufficient exposure and practice, procedural memory can take over for at least some grammatical operations, producing more automatic, native-like performance.

The DP model makes several specific predictions that have received empirical support:

  • Late L2 learners should show weaker procedural advantages for regular morphology (they rely more on memorized forms even for regulars).
  • Immersion and intensive practice should shift L2 grammar processing toward procedural systems over time.
  • Patients with hippocampal damage (declarative memory impaired) should show greater difficulty with irregular past tenses (mice → irregular) while retaining regular morphology (walks, walked → procedural rule).
  • Patients with basal ganglia damage (procedural memory impaired) should show the reverse: regular morphology disrupted, irregular vocabulary relatively intact.

The DP model has been particularly influential in connecting psycholinguistics to clinical populations and bilingualism research, and it offers a principled explanation of why adult L2 learners often permanently show traces of non-native-like grammar processing even with high proficiency.

Key Contributions

  • Declarative/Procedural (DP) model — neurocognitive theory of language processing, distinguishing declarative and procedural memory substrates
  • DP model in L2 acquisition — predictions and evidence about how adult L2 learners rely on declarative memory and gradually proceduralize grammar with experience
  • Clinical SLA predictions — DP model accounts for language deficits in hippocampal and basal ganglia patients

Common Misconceptions

  • The DP model is not a purely SLA theory. It is a general theory of language processing in the brain that has implications for SLA as a special case — the acquisition of a new language system by adults whose memory systems have different developmental histories than children’s.
  • Proceduralization does not mean L2 speakers become identical to L1 speakers. Even with high proficiency, L2 speakers may show different neural activation patterns, reflecting continued partial reliance on declarative memory for grammatical operations that are fully procedural in native speakers.

Related Terms

See Also

Sources

mikeydoes.com