Where we live affects where we can enroll our children in school, how much money we can get for our house, even how we’ll probably vote. Our neighborhood matters.
Psycholinguists think of words as belonging to neighborhoods, too. Phonological neighborhoods are defined as sets of words that differ by a single sound. (There are also orthographic neighborhoods, which differ by a single letter.) Tomb and boom and room are all neighbors, as are tomb and tooth and tune. (Boom and tooth, which differ by two sounds, are not generally considered neighbors, although the distinction between neighbor and non-neighbor is admittedly arbitrary, gerrymandered by researchers for convenience’s sake.)
Why would psycholinguists bother defining phonological neighborhoods? It turns out that a word’s neighborhood size—the number of neighbors that a given word has—can affect the way we understand and use it.
For one, it takes listeners longer to determine what a word is—to understand c-a-t to be cat—when that word has lots of neighbors. That’s because when we hear a word, everything that sounds like that word becomes slightly more accessible in memory. With a large neighborhood at the ready, it is more difficult to eliminate the words that were not said; it’s harder to rule out the possibility that the talker said cut or kit or cot or cad or cap rather than cat. A word like gem, which has fewer neighbors than cat, is simply less confusable, and thus, all else being equal, requires less work to identify.
For talkers, on the other hand, words that dwell in large neighborhoods are easier to say, especially when the words in that neighborhood are said often. That’s because we get lots of practice accessing the sounds (and therefore the words containing the sounds) common to large, high-frequency neighborhoods. Because a word from a popular neighborhood is easier to retrieve from memory, it can be produced more quickly and fluently. Tellingly, stutterers are less likely to have difficulty producing words from these “urban” neighborhoods.
A word’s neighborhood affects not only the ease with which it is said, but also, according to researchers Melissa Baese-Berk (currently at the Basque Center on Cognition, Brain, and Language) and Matthew Goldrick (of Northwestern University)¸ the manner in which we produce it. When using a word with lots of phonological neighbors, we’re likelier to hyperarticulate, making the word sound as distinct (that is, as little like its neighbors) as possible. (Having a neighbor as common as do, for example, is like living next door to someone who goes crazy with the holiday decorations, or regularly mows his lawn naked: wouldn’t we want to distinguish ourselves from them?) This hyperarticulation becomes even more pronounced when the likelihood of mistaking one word for another is at its highest. We’d better speak as clearly as possible if we’re talking about caps or bats and then wish to drag our cat into the conversation.
