Introduction

Pinker and colleagues (Pinker & Ullman, 2002) have argued that morphologically irregular verbs must be stored as full forms in the mental lexicon, while morphologically regular verbs, consisting of a stem + affix, can be computed by rule (STEM + ed). As a result of this representational difference, regular verbs are presumed to undergo decomposition into their constituent morphemes during processing, while irregular verbs do not.

In contrast to Pinker, McClelland and colleagues (McClelland & Patterson, 2002) propose that both regular and irregular verb inflection emerge from a single, integrated mechanism, in which phonological features of the stem are associated with phonological features of the past-tense form for both regulars and irregulars.

If Pinker’s theory is correct, then we should see behavioral and neurophysiological indicators of morphological decomposition for regular, but not for irregular verbs. If a verb is decomposed during processing into STEM + AFFIX, the past tense form should prime its stem as effectively as the stem primes itself.

In a lexical decision task, Stanners et al. (1979) found facilitation for targets primed by themselves, and importantly, an equivalent degree of facilitation for regular verbs primed by their past-tense forms. However, irregular verbs primed by their past tense forms did NOT produce an equivalent degree of facilitation.

Muente et al. (1999) measured ERPs to regular and irregular verbs in a morphological repetition priming task in an attempt to look for decomposition effects. They found that the ERPs to regular English verbs were associated with an N400 reduction when primed by their past tense forms. However, irregular verbs did not show this effect.

However, Muente et al. used a delayed repetition priming paradigm. Delayed repetition priming does not necessarily allow us to distinguish between effects deriving from differences in lexical activation, and those deriving from differences in episodic memory. It is possible that the observed differences in the N400 reduction effect in Muente’s study were due to differences between regular and irregular version in terms of episodic memory rather than lexical organization.

Masked Morphological Priming of Regular and Irregular VerbsMasked Morphological Priming of Regular and Irregular VerbsJoanna Morris Florack1 & Phillip J. Holcomb2

Hampshire College, Amherst, MA1; Tufts University, Medford, MA2

470

480

490

500

510

520

530

540

550

Regular Irregular

Verb Type

Mill

isec

onds

Infinitive

Past Tense

Unrelated

Figure 2: ERPs to Verb Targets preceded by (a) Infinitive (b) Past Tense and (c) Unrelated Primes

Figure 3: Electrode Montage

Figure 4: RTs to Verb Targets preceded by (a) Infinitive (b) Past Tense and (c) Unrelated Primes

This research was supported by HD25889 and HD043251

Discussion

These data show that both regular and irregular verbs show an N400 reduction effect when primed with their past tense forms.

This finding suggests that any differences found between regular and irregular verbs in a delayed morphological priming paradigm may reflect differences in episodic memory rather than differences in lexical representation and processing.

ReferencesMcClelland, J. & Patterson, K. (2002). Rules or connections in past tense inflections: What does the evidence rule out. Trends in Cognitive Sciences, 6(11), 465-472.Muente, T. F., Say, T., Clahsen, H., Schiltz, K., & Kutas, M. (1999). Decomposition of morphologically complex words in English: Evidence from event-related brain potentials. Cognitive Brain Research, 7(3), 241-253.Pinker, S., & Ullman, M. T. (2002). The past and future of the past tense. Trends in Cognitive Sciences, 6(11), 456-463.Stanners, R. F., Neiser, J. J., Hernon, W. P., & Hall, R. (1979). Memory representation for morphologically related words. Journal of Verbal Learning & Verbal Behavior, 18(4), 399-412.

The purpose of this study was to look for neurophysiological indicators of morphological decomposition for regular and irregular verbs using a masked priming paradigm in which primes are not consciously available to the subject. In this case, any observed differences cannot be attributed to episodic memory effects but are likely to reflect true differences in lexical organization.

Methods

• 20 adults (5 men and 15 women). Age Range: 18 to 25(mean 20.7 years)

• Stimuli were 120 regular and 120 irregular past tense and infinitive verb forms, matched for frequency and length (in the infinitive form only).

• Each infinitive form appeared in four conditions (a) primed by itself, (b) primed by its past tense form, (c) primed by an orthographic control that differed in only one letter, and (d) primed by an unrelated item with which it shared no letters.

• In addition to the experimental items, each list contained 480 non-word filler items.

• Subjects were instructed to respond to each target by pressing a button labeled “YES” if the target was a word, and one labeled “NO” if the target was a non-word.

Figure 1: Procedure

Figure 5: ERPs to Verb Targets preceded by (a) Past Tense (b) Orthographic Control and (c) Unrelated Primes

Figure 6: RTs to Verb Targets preceded by (a) Past Tense (b) Orthographic Control and (c) Unrelated Primes

470

480

490

500

510

520

530

540

550

560

Regular Irregular

Verb Type

Mill

isec

onds

Past Tense

Ortho Control

Unrelated

Figure 7: Comparison of Unrelated-Past Tense Difference waves for (a) Regular and (b) Irregular Verb Targets