This study explores the sensitivity of the individuals to the residual gestures remaining after the simplification of consonant clusters. Three sets of target stimuli having full, reduced, and zero alveolar gestures along with the control stimuli were used in a perceptual identification task. The results of the experiment showed that subjects reliably distinguished the three target sets with varying residual gestures from the control. The results also showed that the degree of residual gestures affects the rate of [t] perception by the subjects; however, this was not statistically significant. The results are discussed in the context of different theories of speech perception.
clusters in the words /raeft/ “went”, /duχt/ “sew” and /qaesd/ “intension” are optionally simplified1 in fast/casual speech, resulting in: [raef], [duχ], and [qaes], respectively. The articulatory study conducted on this process in Persian by Falahati (2013) has shown that the gestures of the deleted segments are often still present. More specifically, the findings showed that of the clusters that sounded simplified, some had no 1 The term “simplification” is used here for the acoustic and perceptual consequence of apparent coronal consonant deletion, regardless of whether there is a residual articulatory gesture. alveolar gesture, some had gestural overlap that masked at least some of the acoustic information for [t], and some had reduced alveolar gestures. The current study tests listeners’ sensitivity to these three types of /t/ realizations. 2
Choosing the basic units or building blocks by which the phenomena in a discipline could be explained is fundamentally important. Due to the “complex” nature of language, there is no consensus among linguists as to the nature of this basic unit in the field. The controversy over choosing the building blocks extends to the domain of speech perception where different models have postulated various basic units of processing and storage.
In general, there are two major theoretical
approaches to speech perception: gesturalist
theories versus auditory and exemplar theories.
The two main gestural theories of speech
perception are Motor Theory and Direct Realism
(MT and DR, henceforth). In motor theories, the
intended phonetic gestures of the speaker are
considered to be the objects of speech
perception. These gestures are “represented in
the brain as invariant motor commands that call
for movements of the articulators through certain
linguistically significant configurations”
An intended gesture is produced by a number of muscles that act in concert sometimes ranging over more than one articulator. For instance, constriction needed for producing coronal stops involves the action of the tip/blade of the tongue and the jaw; however, such a constriction is considered one gesture. According to MT, the orchestration among gestures is quite systematic and listeners can use the systematically varying acoustic cues for coronal stops as information to detect the related consonant gestures.
MT assumes a biological link between
perception and production. According to this
perspective both speech perception and speech
production share the same set of invariants and
are governed by auditory principles. “The
motivation for articulatory and coarticulatory
maneuvers is to produce just those acoustic
patterns that fit the language-independent
characteristics of the auditory system”
The other main gestural theory to speech
perception is direct realism. Both DR and MT
share the claim that listeners to speech perceive
vocal tract gestures. However, in DR it is the
phonological gestures of the vocal tract, rather
than the intended gestures, which are the
perceptual objects
In contrast to gestural theories, the auditory
theories assume that speech sounds are perceived
via general cognitive and learning mechanisms.
In this view, speech is not special and listeners
do not perceive gestures. The auditory approach
to perception mainly considers general auditory
mechanisms responsible for perceptual
performance. According to this view, the speech
and nonspeech stimuli do not invoke a special or
speech-specific module. Gestures have no
mediatory role as to the perception of speech
sounds in this approach. Listeners use multiple
imperfect acoustic cues in order to categorize the
complex stimuli with structured variance
Exemplar theories form another approach to speech perception where words and frequentlyused grammatical constructions are represented in memory as large sets of exemplars containing fine phonetic information. Listeners are sensitive to phonetic details existing in the speech signal. In such a speech perception model, a mechanism is needed for gradiently changing the lexical representations over time. In order to do so, the perceptual system must be capable of making fine phonetic distinctions (Johnson 1997).
These different approaches to speech perception have been tested in different studies. Beddor et al., (2013), for example, used eyetracking to assess listeners' use of coarticulatory vowel nasalization as that information unfolded in real time. In the experiment, subjects heard the nasalized vowels with two different time latencies. The prediction was that subjects will fixate on the related image sooner when they hear the nasalized vowel earlier. The results showed that listeners use relevant acoustic cues, which was argued to allow listeners to track the gestural information. Nalon (1992) in an identification task tested whether participants could identify different degrees of velar assimilation. He used four different articulation types called full alveolar, residual alveolar, zero alveolar (i.e., full assimilation to the following velar), and nonalveolar (i.e., velar in underlying representation). The results of his study showed that the participants perceived full alveolar tokens with 100% accuracy with /d/ responses while less than half the tokens with residual alveolar were identified with /d/ responses. In another study, Pisoni showed that the nonspeech analogs of VOT stimuli are perceived categorically. Similar studies like this were taken as evidence against MT which claimed categorical perception as a specific feature of the speech mode of perception.
In this study, I will use three sets of simplified consonant clusters which are auditorily similar but gesturally different. The consonant clusters (i.e., C1C2#) happen in the coda of the words followed by another word which also starts with a consonant, therefore giving us three consonants in a row in an intervocalic environment (i.e., V1C1C2#C3V2). The prediction is that if subjects are sensitive, they should have different judgment for the stimuli. The stimuli set with no coronal gesture is expected to show the same pattern as the control (with zero coronal gesture in the underlying representation). The stimuli with overlapped gestures and reduced gestures are predicted to show a pattern different both from control and the stimuli with zero residual gestures. The following section introduces the methodology of the study. 3 3.1
Thirty-two Persian-speaking students from the Università di Pisa and Sant’Anna, seventeen females fifteen males, aged 18-38 participated in this study. The results of eight of them are not considered for analysis because they reported to be bilinguals and mainly used a language rather than Persian at home or with their close friends. This resulted in twenty-four, twelve females twelve males. None of them reported any hearing problem. 3.2
Three sets of target words varying in only the degree/amount of alveolar residual gestures and one control stimuli set were used in the experiment. The three target categories are mainly the same except for the degree of alveolar residual gestures. Target Full_G category has full coronal gesture but has overlap hence marked with two superscript [tt]. Target Partial_G category has partial residual gesture marked via superscript [t] whereas Target Zero_G has no gestural leftover. The stimuli in the control are used as the baseline since they don’t have any underlying coronal stop in the coda position of the first word. Some examples of the target and control words are given below:
Target Full_G: [aeχtt kɑ], [aeftt bae], [uftt bɑ]
Control: [aeχ ke], [aef bae], [uf bɑ] The four sets of target and control nonwords presented above are the excised tokens taken from the full words presented below:
kɑr/ “hard-working”, /naeft baerɑje/ “oil for”, /kuft bɑʃeh/ “be cheap” Control: /naeχ ke/ “thread that”, /saef baerɑje/, “cue for” / maeruf bɑʃeh / “be famous” 3.3
All the participants listened to forty stimuli (10 stimuli in each category) with eight repetitions. (total of 320 tokens) in a sound booth located at the linguistics laboratory in Scuola Normale Superiore. The software Presentation was used to present the stimuli to the listeners as an identification task. The participants were asked to listen very carefully and decide as quickly as possible whether it is likely that there has been a [t] at the end of the first part of each stimuli. For each stimulus, the participants were asked to press either the green or the blue button on a Cedrus response pad. On the screen of a computer, listeners could also see “T” or “NO T” corresponding to the response buttons. The stimuli were shuffled and presented in blocks in a way that participants could either begin by hearing all the tokens with [f] or [χ]. They also had the choice of taking a break after listening to every 80 tokens. All the participants received a short training before the start of the experiment. The following section contains the results of the study. 4
The main goal of this study is to test listeners’ sensitivity to different degrees of residual gestures remaining after the simplification of consonant clusters. The response type and reaction time are the dependent variables in this study; however, only the results related to response type are presented here. Figure 1 below shows the perception rate of [t] by all subjects across the four conditions. According to this, the subjects show the highest rate of [t] perception in tokens with full alveolar gesture (i.e., 59.69%) and the lowest for the ones in the control (i.e., 36.09%). The condition with partial alveolar gestures shows the rate of 56.20% which is very close to the full condition. The stimuli in zero alveolar condition show an intermediate level between the control and the other two target conditions with the rate of 49.84%. This shows almost a similar pattern between the two target conditions with full and partial gestures, an intermediate situation for the target condition with zero gesture, and a pattern for the control which is different from the three target conditions.
The rate of [t] perception for the
mean subject In order to examine the relation between the two categorical variables in the study, namely the response type and stimuli condition, a Pearson chi-square test was run. The null hypothesis is that there is no relation in the [t] perception and the four conditions in the study. The results of the test with [t] perception as the dependent variable found significant main effect of conditions χ2 (3, N = 960) = 46.2, p < 0.001. This shows that there is a significant relation between the stimuli conditions and response type. In order to determine whether the difference in the perception of [t] across four categories is really significant or it is due to chance variation, a column proportions test was performed. This test uses z-test to make the comparisons. The result showed that the perception of [t] in the control was significantly different from the all target categories. The next section presents the discussion and concluding remarks of the study.
This research investigated listeners’ sensitivity to three types of /t/ realizations as target and compared the results with the control. The target categories included simplified consonant clusters with full, partial, and zero alveolar gestures. The stimuli used as the baseline in the control had no alveolar gesture in the underlying form. The general results of the study showed that subjects reliably distinguished the three target sets with varying residual gestures from the control. This could be due to more similarity in tongue configuration in realizing these varying degrees of coronal stop articulation compared to the control condition where there is no alveolar gesture in the underlying form. Any articulatory modification is expected to trigger acoustic changes. The acoustic results of the stimuli used in this study by Falahati (2013) showed no significant difference between the simplified tokens (i.e., the three target sets with varying degrees of residual gestures labeled all together as simplified) and control tokens. The acoustic parameters used in the analysis were V1 duration, consonant clusters duration, and formant transitions. Despite of the fact that the results did not show any significant difference between simplified and control conditions, the duration of V1 and consonant clusters in the simplified condition was always higher than the control condition. It could be the case that these acoustic cues, although not very strong, are enough for human’s auditory system to trigger the presence of a segment.
The results of the current study also showed that participants perceived almost 36% of the tokens with no underlying coronal stop as having [t]. This is very similar to the results of the study reported by Nalon (1992) where 20% of the control nonalveolar tokens were perceived as having [d]. In his study, however, the control tokens showed similar pattern to that of the target with zero alveolar (i.e., full assimilation). He attributes this to both subjects’ natural language experience as well as the inherent ambiguity in the stimuli. He states that subjects are “willing to “undo” its effects” and therefore, in the case of the current study, report coronal stops even where there is no evidence for them.
The results of our study also showed that participants perceived more [t] in the tokens with full and partial alveolar gestures compared to the ones with zero alveolar gestures. The difference between the three categories, however, did not reach the significance level. Such result could shed more light on the theories of speech perception discussed earlier in this paper. In order to discuss this issue, first we need to further explore the nature of the three categories in the target stimuli. From the three groups in the target stimuli, one group categorically had no alveolar gesture while the other two had different degrees of the gesture either as a result of overlap or reduction. We argue that the gradient gestural reduction and overlap are due to lowlevel phonetic and mechanical reasons while the categorical deletion, which results in tokens with zero gestures, is caused by the cognitive system. In the former groups, speakers neither intend to reduce nor plan to overlap gestures while the latter process is intended by the speaker.
According to MT and DR, listeners’ target in speech perception is the intended or phonological gestures. Therefore, the overlapped and reduced stimuli should show different perceptual pattern compared to the stimuli with no residual gesture. The results in this study did not show a striking difference between these three target sets. The existence of acoustic cues pertaining to the presence of gestures is a prerequisite to their perception by the listener. If distinguishing acoustic details could be found between these three categories, then this would not support the gesturalist approach to speech perception. However, with the current results, such a claim cannot be made. Further acoustic analysis between these three target sets is needed to examine this idea further.
The findings in our experiment could be best explained by referring to exemplar models of speech perception. In such models, the lexical representations of words change in a gradient way over time. This is due to the nature of some phonological processes in languages which are not categorical. According to this view, the perceptual mechanism is capable to make fine phonetic distinctions. However, it is the mapping between the gradient stimuli and the auditory system which fails and does not result in nonvariant forms.
The lack of such a one-to-one mapping will bring variation across subjects in the speech community. The degree of such variation is determined by the amount of individual’s exposure to the specific variants. A closer look at the results for individual subjects showed that all twenty-four participants in the study could fall into three or four dominant patterns based on their perception of [t]. The variation across individuals regarding speech perception could be a good source of information for the specialists in the field. Moreover, the degree to which an individual’s speech production could map to his/her perception is an interesting topic which remains to be explored.
We are very grateful to Patrice Beddor for her comments and suggestions on this study.
Alvin M. Liberman and Ignatius G. Mattingly. 1985.
The motor theory of speech perception revised.
Cognition, 21: 1-36.
Francis Nalon, 1992. The descriptive role of segments: evidence from assimilation. In G. J. Docherty and R. Ladd (eds.), Papers in Laboratory Phonology IV, 261-289. Cambridge: Cambridge University Press.