Web Activity 8.2 Neighborhood density

Dense or sparse neighborhoods?

Consider the words below. How many phonological neighbors can you find for each (i.e., other words that share all but one of that word’s phonemes)? Be sure to pay attention to the sounds (phonemes) of the word, rather than the symbols used to spell it.

• kill
• warmth
• glance
• wend
• wait
• sleight
• slouch
• type
• could
• scourge
• thing
• sketch
• swerve

Now try to come up with your own examples of words that have few phonological neighbors (fewer than eight) and words that have many neighbors (more than fifteen). Aim for a list of twenty words, with ten high- and ten low-neighborhood-density items.

Other variables that affect word recognition

Tests of spoken word recognition show that words that come from dense phonological neighborhoods take longer to recognize than words that have very few phonological neighbors. This suggests that words that sound similar to the spoken word compete with the target word representations. So on the whole, you would predict that the high-density words you identified in the first part of this exercise would take longer to recognize than the low-density words. But other factors can affect the recognition time of spoken words. For example, the word scourge has a small number of sound-alike neighbors while thing has many; but the word thing is far more common than scourge. Because of this, it would not be surprising to find that subjects take less time to recognize thing than scourge. Experiments that fail to take this into account could easily miss the contribution of neighborhood density on word recognition.

Consider the example words from the list above. Are there other factors inherent to these words that you think might affect recognition times? Venture some hypotheses about their possible effects.

A resource for accessing neighborhood densities and word frequencies

Fortunately, there are a number of handy tools available to researchers who design experiments in word recognition. One of these is the Irvine Phonotactic Online Dictionary (Vaden et al., 2009; http://www.iphod.com), which allows researchers to quickly assess a number of properties of English words, including their frequencies and their phonological neighborhoods as computed from a database of more than 54,000 words. You can use this publicly-accessible tool yourself to evaluate the words you came up with in the first part. To do this:

• Go to the website http://www.iphod.com/search/V2ListWords.html
• In part 1, type your target word into the search field.
• In part 2, check the box marked “Unweighted” next to “Density” and “SFreq” next to “SUBTLEXus word frequency measures” to get a basic frequency count and number of phonological neighbors.
• In part 3, click the button marked “Search IPhOD Version 2.0.”

Were the words you chose for the high- and low-density lists well-balanced for word frequency and number of phonemes? Play around with the IPhOD search tool to try to come up with a better list of experimental items. Beside each item, list the number of phonemes it contains, the size of the phonological neighborhood, and its word frequency count.

Reference

Vaden, K. I., Halpin, H. R., Hickok, G. S. (2009) Irvine Phonotactic Online Dictionary, Version 2.0. [Data file]. Available from http://www.iphod.com.