LECTURE 2.4 – Learning and Processing the Past Tense
There is a relatively fixed pattern with which children acquire past tense morphemes...
o “-ing,” plural “s,” possessive “s,” third person “s,” irregular past tense, regular past
The question is do children learn these rules, or are they memorizing?
Berko (1958) answered this question with his classic Wug Test. He invented a fictitious creature
called a Wug. He displayed one on paper with text that said “this is a wug.” Then in another
picture he had two of them with text that said “add another, now there are two ___.”
Children filled in the missing word as “wugs” which conclusively proves that children do learn
Children as young as 3 exhibit this trait.
What about irregular verbs? English has about 180 irregular verbs whose past tense form is
completely arbitrary. Therefore children must memorize these.
Kids effectively learn irregular verbs before regular ones however there is a period of relapse
when all of a sudden their performance in using irregulars declines and they slowly work their
way back up. This is an example of a U-Shaped Learning Curve.
Steve Pinker explains this u-shaped learning; he contends that children first learn irregulars, but
when they discover the “-ed” rule they over-generalize it to irregular verbs. They then have to
undergo the process of relearning irregular verbs.
Pinker argues for something called the dual route theory.
Dual Route Theory...
o This is a modular theory.
o It argues that the “-ed” rule is encoded through a separate route than the regulars.
o Claims that the Rule Route is where we produce the regular past tense verbs and the
Irregular Route leads to LTM where we store the irregulars.
Now we will consider some interesting observations and see if we can later apply the Dual Route
Theory to them.
It seems that the regular ending is treated as the default for both new words (i.e. blogged,
unfriended, etc.) and non-sense words in studies (i.e. taffed, blorped, leffed, etc.).