Presentations

Abstract

Subject-verb agreement is one of the most common
grammatical encoding operations in language
production. In many languages, morphological
inflection on verbs code for the number of the head
noun of a subject phrase (e.g., The key to the cabinets
is rusty). Despite the relative ease with which subjectverb
agreement is accomplished, people sometimes
make agreement errors (e.g., The key to the cabinets
are rusty). Such errors offer a window into the early
stages of production planning. Agreement errors are
influenced by both syntactic and semantic factors, and
are more likely to occur when a sentence contains either
conceptual or syntactic number mismatches. Little
is known about the timecourse of these influences,
however, and some controversy exists as to whether
they are independent. The current study was designed
to address these two issues using EEG. Semantic and
syntactic factors influencing number mismatch were
factorially-manipulated in a forced-choice sentence
completion paradigm. To avoid EEG artifact associated
with speaking, participants (N=20) were presented with
a noun-phrase, and pressed a button to indicate which
version of the verb ‘to be’ (is/are) should continue
the sentence. Semantic number was manipulated
using preambles that were semantically-integrated or
unintegrated. Semantic integration refers to the semantic
relationship between nouns in a noun-phrase, with
integrated items promoting conceptual-singularity.
The syntactic manipulation was the number (singular/
plural) of the local noun preceding the decision. This
led to preambles such as “The pizza with the yummy
topping(s)... “ (integated) vs. “The pizza with the tasty
bevarage(s)...” (unintegrated). Behavioral results showed
effects of both Local Noun Number and Semantic
Integration, with more errors and longer reaction times
occurring in the mismatching conditions (i.e., plural
local nouns; unintegrated subject phrases). Classic ERP
analyses locked to the local noun (0-700 ms) and to the
time preceding the response (-600 to 0 ms) showed no
systematic differences between conditions. Despite this
result, we assessed whether difference might emerge
using multivariate pattern analysis (MVPA). Using the
same epochs as above, support-vector machines with a
radial basis function were trained on the single-trial level
to classify the difference between Local Noun Number
and Semantic Integration conditions across time and
channels. Results revealed that both conditions could
be reliably classified at the single subject level, and
that classification accuracy was strongest in the epoch
preceding the response. Classification accuracy was
at chance when a classifier trained to dissociate Local
Noun Number was used to predict Semantic Integration
(and vice versa), providing some evidence of the
independence of the two effects. Significant inter-subject
variability was present in the channels and time-points
that were critical for classification, but earlier timepoints
were more often important for classifying Local Noun
Number than Semantic Integration. One result of this
variability is classification performed across subjects was
at chance, which may explain the failure to find standard
ERP effects. This study thus provides an important first
test of semantic and syntactic influences on subject-verb
agreement with EEG, and demonstrates that where
classic ERP analyses fail, MVPA can reliably distinguish
differences at the neurophysiological level.

Abstract

Prominent theories of predictive language processing assume that language production processes are used to anticipate upcoming linguistic input during comprehension (Dell & Chang, 2014; Pickering & Garrod, 2013). Here, we explore the converse case: Does a task set including production in addition to comprehension encourage prediction, compared to a task only including comprehension? To test this hypothesis, we conducted a cross-modal naming experiment (Experiment 1) including an object naming task and a self-paced reading experiment (Experiment 2) that did not include overt production. We used the same predictable (N = 40) and non-predictable (N = 40) sentences in both experiments. The sentences consisted of a fixed agent, a transitive verb and a predictable or non-predictable target word (The man drinks a beer vs. The man buys a beer). Most of the empirical work on prediction used sentences in which the target words were highly predictable (often with a mean cloze probability > .8) and thus it is little surprising that participants engaged in predictive language processing very easily. In the current sentences, the mean cloze probability in the predictable sentences was .39 (ranging from .06 to .8; zero in the non-predictable sentences). If comprehenders are more likely to engage in predictive processing when the task set involves production, we should observe more pronounced effects of prediction in Experiment 1 as compared to Experiment 2. If production does not enhance prediction, we should observe similar effects of prediction in both experiments. In Experiment 1, participants (N = 54) listened to recordings of the sentences which ended right before the spoken target word. Coinciding with the end of the playback, a picture of the target word was shown which the participants were asked to name as fast as possible. Analyses of their naming latencies revealed a statistically significant naming advantage of 106 ms on predictable over non-predictable trials. Moreover, we found that the objects’ naming advantage was predicted by the target words’ cloze probability in the sentences (r = .411, p = .016). In Experiment 2, the same sentences were used in a self-paced reading experiment. To allow for testing of potential spill-over effects, we added a neutral prepositional phrase (buys a beer from the bar keeper/drinks a beer from the shop) to each sentence. Participants (N = 54) read the sentences word-by-word, advancing by pushing the space bar. On 30% of the trials, comprehension questions were used to keep up participants' focus on comprehending the sentences. Analyses of participants’ target and post-target reading times revealed numerical advantages of 6 ms and 20 ms, respectively, in the predictable as compared to the non-predictable condition. However, in both cases, this difference was not statistically reliable (t = .757, t = 1.43) and the significant positive correlation between an item’s naming advantage and its cloze probability as seen in Experiment 1 was absent (r = .037, p = .822). Importantly, the analysis of participants' responses to the comprehension questions, showed that they understood the sentences (mean accuracy = 93%). To conclude, although both experiments used the same sentences, we observed effects of prediction only when the task included production. In Experiment 2, no evidence for anticipation was found although participants clearly understood the sentences and the method has previously been shown to be sensitive to measure prediction effects (Van Berkum et al., 2005). Our results fit with a recent study by Gollan et al. (2011) who found only a small processing advantage of predictive over non-predictive sentences in reading (using highly predictable sentences with a cloze probability > . 87) but a strong prediction effect when participants read the same sentences and carried out an additional object naming task (see also Griffin & Bock, 1998). Taken together, the studies suggest that the comprehenders' task set exerts a powerful influence on the likelihood and magnitude of predictive language processing. When the task set involves language production, as is often the case in natural conversation, comprehenders might engage in prediction to a stronger degree than in pure comprehension tasks. Being able to predict words another person is about to say might optimize the comprehension process and enable smooth turn-taking.

Abstract

If message and sentence planning are closely linked processes, planning scope may vary depending on what speakers want to say and how they say it. We compared speakers’ gaze pattern to pictures in displays eliciting sentences like “The lion and the tiger are above the basket” when speakers were a) more familiar or less familiar with the spatial layout of these displays, and b) more familiar or less familiar with the phrasal structures used in these sentences. Familiarity with spatial layout was induced by presenting prime trials with a similar or dissimilar layout of pictures (“The bell and the nail are above/below the crutch”) before the target trial, and familiarity with sentence structure was manipulated via structural priming (prime trials elicited sentences like “The bell and the nail are above the crutch” or “The bell is above the nail and the crutch”). When describing pictures on target trials, speakers looked earlier at the second object (tiger) when they were familiar with both the spatial layout and sentence structure, but speech onsets were reduced (structural priming) only when both spatial layout and sentence structure were repeated. The results show that linguistic planning is facilitated by congruence between message-level and sentence-level structure.

Abstract

Studies of lexical access in speech planning often use priming or interference paradigms, where a target picture is combined with a written prime or distracter word. A difficulty in interpreting the results of studies using interference paradigms with clearly visible distracters is that effects arising during lexical access cannot be distinguished from effects arising during self-monitoring. A difficulty with using masked priming paradigms is that the effects tend to be small and fragile. We report a series of picture naming experiments using both the conventional masked priming procedure and the sandwich priming procedure first used in lexical decision experiments by Lupker and masked prime the participants are briefly (i.e., 33 ms) presented the name of the target picture. Although neither categorically nor phonologically related primes significantly affected picture naming in the traditional masked priming experiments, in the sandwich priming experiments: (a) categorically related primes (e.g. “dog-cat”) interfered more than unrelated distracters with picture naming and (b) phonologically related primes (“mat-cat”) facilitated picture naming. The theoretical implications of these findings will be discussed. Lupker, S.J., & Davis, C.J. (2009). Sandwich priming: A method for overcoming the limitations of masked priming by reducing lexical competition effects. Journal of Experimental Psychology: Learning, Memory, & Cognition, 35, 618-639.

Abstract

Models of the mental representation of morphologically complex words traditionally fall into one of two categories, Single-Route or Dual-Route models. The former further distinguish between Full-Listing (e.g. Butterworth, 1983) and Decomposition (e.g. Taft & Forster, 1976), while the latter assume different systems governing the access of mono- vs. polymorphemic words (e.g. Pinker & Prince, 1994; Pinker & Ullman, 2002). One of the main arguments against decomposition and continuous online computations is the cognitive resources this process would require. Turning this reasoning around, taxing someone's working memory capacities should then uniquely affect the computation of bimorphemic verb forms. We tested this hypothesis on 48 Dutch native speakers with a lexical decision task, comparing reaction times for Dutch regular past tense forms to frequency-matched irregular past tense forms, both under low and under high cognitive load. We found that frequency influenced reactions to monomorphemic but not to bimorphemic forms (F(1, 47) = 4.734, p = .035), favoring a listing account for the former but a computational procedure for the latter forms. This interaction, however, was present only for a certain group of people (F(1, 23) = 6.279, p = .02), namely those whose reaction times were hardly affected by the load manipulation and who thus may be thought of as having larger working memory capacities. On the other hand, participants who showed a strong load effect had no interaction between number of morphemes and frequency (F(1, 23) = .575, ns), indicating that they process monomorphemic and bimorphemic forms in a similar manner. It seems that cognitive capacities influence the storage of and access to polymorphemic verb forms. While people with greater working memory skills use these resources to compute morphologically complex inflections on-line, people with smaller cognitive capacities seem to rely on a list-like storage for bimorphemic forms as well.

Abstract

Speakers clearly differ in how quickly they can retrieve words from the mental lexicon, but little is known about the sources of this variability. The present study investigated the relationship between speakers’ executive control abilities and their speed of picture naming. In two experiments, adult speakers of British English named line drawings of objects and actions. Three main components of executive control - updating, shifting of attention, and inhibiting - were assessed using the operation-span, number-letter shifting, and stop-signal task, respectively (see Myake et al.,2000 ). Reaction times (RT) to action and object pictures were highly correlated. Ex-Gaussian analyses of the RT distributions showed that the speakers’ updating scores correlated with the tau parameter of the RT distributions, i.e. predicted the proportions of slow responses in action and object naming. The inhibiting scores correlated with the mean RTs, whereas the scores obtained in the number-letter shifting task were uncorrelated to the RTs. These results indicate that the executive control abilities of updating and inhibiting contribute to the speed of naming objects and actions. Theories of word production may require modification to take account of these findings.

Abstract

a bitter rain – two silences beneath the one umbrella Is the connotative meaning of texts readily available or is it gleaned at an extra cognitive cost? The eye-movements of 31 English native speakers (10 male, mean age 21 years) were recorded while reading 24 haiku, 12 in the original/symbolic version, and 12 in a modified version where the most symbolic word (the keyword ‘bitter’ in the example) was replaced by a more literal word (‘loud’) reducing the text’s symbolic purport. The effects of keyword substitutions were measured globally, comparing total reading times for the two haiku types, and locally, examining the first pass gaze durations and dwell times on a word closely connected to the keyword, the referent ‘silences’, and on the last word, ‘umbrella’, to examine wrap-up processes. First pass duration showed no effects of the substitution. However, dwell time on referents and last-word regions, and total reading time were significantly longer for the original than for the altered haiku, suggesting that the connotative meaning of the texts was not available immediately but only through re-reading of the texts. The implications of these findings are discussed in relation to the literature on the processing of inferences in symbolic texts.

Abstract

a bitter rain – two silences beneath the one umbrella Is the connotative meaning of a text readily available or is it accessed only after re-examining a text? Thirtyone English native speakers (10 male, mean age 21) read 24 haiku, 12 in their original/symbolic version and 12 in a version in which the most symbolic word (the keyword "bitter" in the example above) was replaced by a more literal word ("loud"), reducing the text’s symbolic purport. Participants' eye movements were recorded using the eye-tracker EyeLink 1000. The effect of the word substitution was measured globally, by comparing the total reading times for the two haiku types, and locally, by examining first pass duration and dwell time on a word closely connected to the keyword (the referent, "silences") and on the last word ("umbrella"), as this might show wrap-up processes. First pass durations showed no effects of the substitution. However, total reading time and dwell time on both the referent and the last-word regions were significantly longer for haiku with the original keyword than for haiku with the altered keyword. These findings suggest that the texts’ connotative meaning was not available immediately but only through re-reading of the texts.

Abstract

When children learn their native language, they have to deal with a confusing array of dependencies between various elements in an utterance. Some of these dependencies may be adjacent to one another whereas others can be separated by considerable intervening material. In this study, we investigate whether both types of dependencies can be learned together, similarly to the task facing young children. Statistical learning of adjacent dependencies (probability = .17) and non-adjacent dependencies (probability = 1.0) was assessed in two experiments using a modified serial-reaction-time task. The results showed (i) increasing online sensitivity to both dependency types during training, (ii) better nonadjacency than adjacency learning, and (iii) nonadjacency learning being highly correlated with adjacency learning, suggesting that adjacency and non-adjacency learning can occur in parallel and might be subserved by a common statistical learning mechanism. An overnight break between two training sessions helped the online learning performance of slower learners to approach that of faster learners, but the same amount of training without such a break (a 15-min interval) did not, suggesting that memory consolidation may play a role in statistical learning of complex statistical patterns, especially for slower learners.