January 01, 2007

Girls, boys, and verb forms


The 12/22/06 issue of Science has a brief report (in Constance Holden's "Random Samples", p. 1845) about research on the learning of verb forms by children, research suggesting that "boys and girls employ slightly different strategies in language-learning".  Obviously a topic of interest here at Language Log Plaza.  The description in Science was surprisingly hard for me to follow, and I'm a morphologist (among other things), so I wondered whether non-linguist readers could figure things out.  As it turns out, the abstract of the article in question -- Joshua K. Hartshorne and Michael T. Ullman, "Why girls say 'holded' more than boys", Developmental Psychology 9.1.21-32 (January 2006) -- is significantly clearer than the Science account.


My confidence in the Science account was shaken when I noticed that it said the research was reported in the November issue of Developmental Science, though it turns out that the paper appeared in last January's issue (and was available on-line in December 2005), so it's not exactly late-breaking news.

In any case, the Science report (titled "He Said, She Said") explains:

As tots learn new words, they tend to "overregularize" verbs--that is, apply the past tense "-ed" even to irregular ones, saying "holded" instead of "held," for example.

To see whether the sexes differ, Michael Ullman and colleagues [in psychology at Georgetown] analyzed transcripts of utterances by 25 children--10 girls and 15 boys--between the ages of 2 and 5.  Because girls learn words faster and are more verbally fluent than boys, Ullman's team suspected that the girls would be better at irregular verbs.  But they found that the girls overregularized more than three times as often as did the boys.

Fine so far.  But the reference to overregularization (an entirely standard piece of terminology, by the way) is likely to suggest to readers that RULES are central to the phenomenon.  That is, at this point the reader is probably thinking that in  producing verb forms, girls apply rules much more than boys do; boys, presumably, produce forms they've memorized.  A reader who goes down this path will be mightily puzzled by what comes next:

By comparing how the tots handled words that sound similar, the researchers claim they could distinguish whether the children were using associative strategies or following rules in deciding verb endings.  When boys overregularize, they are more likely to use rule-governed, or "procedural," memory...  But girls are more likely to go with associations--because the past tense of blink is blinked, sink would become "sinked."

Whoa!  How to interpret this?  It sounds backwards.  And where did the associative strategies come from?

What we needed up front is something about how overregularization could happen in two different ways: by using an internalized rule (in "procedural memory") to compose a form "from scratch"; or by analogizing from forms in memory ("declarative memory") on the basis of similarities -- in particular, similarities in pronunciation -- between lexical items.  This would connect the Georgetown research to controversies in psycholinguistics about the roles of these two types of memory in the production of language.  And it would help in avoiding misunderstanding.

The Hartshorne and Muller paper argues that girls use the analogizing strategy more than boys.  Here's the abstract:

Women are better than men at verbal memory tasks, such as remembering word lists. These tasks depend on declarative memory. The declarative/procedural model of language, which posits that the lexicon of stored words is part of declarative memory, while grammatical composition of complex forms depends on procedural memory, predicts a female superiority in aspects of lexical memory. Other neurocognitive models of language have not made this prediction. Here we examine the prediction in past-tense over-regularizations (e.g. holded) produced by children. We expected that girls would remember irregular past-tense forms (held) better than boys, and thus would over-regularize less. To our surprise, girls over-regularized far more than boys. We investigated potential explanations for this sex difference. Analyses showed that in girls but not boys, over-regularization rates correlated with measures of the number of similar-sounding regulars (folded, molded). This sex difference in phonological neighborhood effects is taken to suggest that girls tend to produce over-regularizations in associative lexical memory, generalizing over stored neighboring regulars, while boys are more likely to depend upon rule-governed affixation (hold+-ed). The finding is consistent with the hypothesis that, likely due to their superior lexical abilities, females tend to retrieve from memory complex forms (walked) that men generally compose with the grammatical system (walk+-ed). The results suggest that sex may be an important factor in the acquisition and computation of language.

The paper is a preliminary opening-up of research in this area.  As Steve Pinker notes in his comments on it in Science, the number of subjects is small, and as the authors note in their conclusion, there is probably considerable individual variation within the groups studied.  Science also cites Pinker's cautious characterization of the research as showing that "males and females sometimes use 'different mixtures of underlying processes' to arrive at the same results" (note "sometimes" and "different mixtures" -- "slightly different mixtures" earlier in the Science article).  Everybody has avoided seeing starkly drawn male-female differences in this research.  Let's hope the mainstream media and pop science writers do as well.

zwicky at-sign csli period stanford period edu

Posted by Arnold Zwicky at January 1, 2007 06:05 PM