Phonological code retrieval during picture naming: Influence of consonant class

Abstract

Investigations of the time course of various stages of lexical processing have indicated either early or late onset of brain activation for phonological code retrieval. The basis of the differential findings is unclear, but factors related to segmental phonology appear to be part of it. The purpose of the present study was to determine whether phonological encoding is influenced by consonant type. Undergraduate students were presented pictures of common and familiar objects to name. Each picture label had an initial liquid (/l/, /ɹ/) or a stop (/b/, /d/) consonant. Accuracy of picture naming was high and comparable for the two stimulus sets. However, words beginning with liquids elicited larger N2 ERP responses than did those with initial stops. Cluster permutation analysis indicated that the ERP responses elicited by words in the two stimulus sets differed between 293 ms and 371 ms post picture onset. These findings point to a late onset of phonological code retrieval. They have implications for segmental phonology and/or motor planning and execution of speech.

Keywords: Adult psycholinguistics; Distinctive features; Event-related potentials (ERP); Lexical processing; Picture naming; Segmental phonology.

Figure 1

ERP waveforms for the 42 electrodes from which amplitude measures were made.

Figure 2

View of three mid-line electrodes (FCz, Cz, CPz) showing P2-N2-P3 peaks in the ERP waveform.

Figure 3

Raster diagram illustrating differences between ERPs elicited by object labels beginning with liquids and stops. Black rectangles indicate electrodes/time points in which the ERPs for liquids are more negative than those for stops. The color scale indicates the size of the t-test result: The darker the color, the higher the level of significance. Note that the electrodes are organized along the y-axis somewhat topographically. Electrodes on the left and right sides of the head are grouped on the figure’s top and bottom, respectively; midline electrodes are shown in the middle. Within those three groupings, y-axis top-to-bottom corresponds to scalp anterior-to-posterior. One outcome was apparent: a broadly distributed effect from 293–371 ms, signifying the time period during which object labels beginning with liquids elicited larger N2 responses than did those with initial stops. The smallest significant t-score was: t(33) = −2.037, p < .03, which had a test-wise alpha level of 0.0497.

Figure 4

Mid-line electrode responses to words with initial stops and liquids that had high (A) and low (B) frequencies. In the high frequency analysis, cluster-based permutation tests revealed no significant consonant class effect between 150 and 400 ms, although visual inspection indicated that words with initial liquids elicited larger N2 responses than did stops. By contrast, significant and extended consonant class effects for the low frequency words beginning with liquids were indicated. These words elicited significantly larger N2 responses across nearly the entire 250 ms analysis window. Responses to words with initial liquid and stop consonants were also combined into the third set of waveforms to examine overall frequency differences (C). The ERP waveforms revealed no statistical differences between high and low frequency words.

Figure 5

Raster diagram illustrating differences between ERPs elicited by low frequency words with initial liquids and stops. Two broadly distributed significant clusters were identified from 187 to 254 ms and from 280 to 400 ms. During these periods, words beginning with liquids elicited larger N2 responses than did their counterparts with initial stops. The smallest significant t-score was: t(33) = −2.035, p < .02, which had a test-wise alpha level of 0.0499.

Figure 6

Mid-line electrode responses to one-syllable and multisyllabic words beginning with (A) stop and (B) liquid consonants. Some variation was observed in the ERP waveform for liquid and stop consonants. Cluster-based permutation tests indicated no significant syllable-length effects during the proposed phonological processing time windows (150–400 ms). Due to the small number of trials, responses to one-syllable and multisyllabic words with initial liquid and stop consonants were also combined into the third set of waveforms (C). The combined ERP waveforms had no statistical differences.

Figure 7

Experimental procedure. Participants were presented single color cartoon pictures, one at a time, on a computer screen. They were asked to overtly name the objects as quickly as possible. Verbal responses were recorded by a voice-activated microphone that also recorded naming response time.

Figure 8

Nine electrode groups selected by scalp location based on Strijkers et al. (2010).