Internet-based language production research with overt articulation: Proof of concept, challenges, and practical advice

Language production experiments with overt articulation have thus far only scarcely been conducted online, mostly due to technical difficulties related to measuring voice onset latencies. Especially the poor audiovisual synchrony in web experiments (Bridges et al. 2020) is a challenge to time-locking stimuli and participants’ spoken responses. We tested the viability of conducting language production experiments with overt articulation in online settings using the picture–word interference paradigm – a classic task in language production research. In three pre-registered experiments (N = 48 each), participants named object pictures while ignoring visually superimposed distractor words. We implemented a custom voice recording option in two different web experiment builders and recorded naming responses in audio files. From these stimulus-locked audio files, we extracted voice onset latencies offline. In a control task, participants classified the last letter of a picture name as a vowel or consonant via button-press, a task that shows comparable semantic interference effects. We expected slower responses when picture and distractor word were semantically related compared to unrelated, independently of task. This semantic interference effect is robust, but relatively small. It should therefore crucially depend on precise timing. We replicated this effect in an online setting, both for button-press and overt naming responses, providing a proof of concept that naming latency – a key dependent variable in language production research – can be reliably measured in online experiments. We discuss challenges for online language production research and suggestions of how to overcome them. The scripts for the online implementation are made available.

Distributional properties of semantic interference in picture naming: Bayesian meta-analyses

Studies of word production often make use of picture-naming tasks, including the picture-word-interference task. In this task, participants name pictures with superimposed distractor words. They typically need more time to name pictures when the distractor word is semantically related to the picture than when it is unrelated (the semantic interference effect). The present study examines the distributional properties of this effect in a series of Bayesian meta-analyses. Meta-analytic estimates of the semantic interference effect first show that the effect is present throughout the reaction time distribution and that it increases throughout the distribution. Second, we find a correlation between a participant’s mean semantic interference effect and the change in the effect in the tail of the reaction time distribution, which has been argued to reflect the involvement of selective inhibition in the naming task. Finally, we show with simulated data that this correlation emerges even when no inhibition is used to generate the data, which suggests that inhibition is not needed to explain this relationship.

Hearing "Birch" Hampers Saying "Duck" – An ERP Study on Phonological Interference in Immediate and Delayed Word Production

When speakers name a picture (e.g., “duck”), a distractor word phonologically related to an alternative name (e.g., “birch” related to “bird”) slows down naming responses compared to an unrelated distractor word. This interference effect obtained with the picture-word interference (PWI) task is assumed to reflect the phonological co-activation of close semantic competitors and is critical for evaluating contemporary models of word production. In the present study, we determined the event-related brain potential (ERP) signature of this effect in an immediate and a delayed version of the PWI task. ERPs revealed a differential processing of related and unrelated distractors: an early (305 – 436 ms) and a late (537 – 713 ms) negativity for related as compared with unrelated distractors. In the behavioral data, the interference effect was only found in immediate naming, while its ERP signature was also present in delayed naming. The time window of the earlier ERP effect suggests that the behavioral interference effect indeed emerges at a phonological processing level, while the functional significance of the later ERP effect is as yet not clear. The finding of a robust ERP correlate of phonological co-activation might facilitate future research on lexical processing in word production.

Failure to show robust noun-referent associations in German-learning infants

Work with the looking-while-listening paradigm suggested that six-month-old English-learning infants associated several common nouns with pictures of their referents. This was evidenced by systematic fixations towards the named target picture (while one distractor picture was present). However, Norwegian-learning infants did not systematically fixate the target pictures until they were 8- to 9-months old, suggesting a cross-linguistic difference in the onset of noun comprehension. Moreover, their success in this task appeared to be modulated by aspects of stimulus pairing, specifically frequency differences between target and distractor: High (resp. low) frequent targets attracted more fixations if they were paired with low (resp. high) frequent distractors. In the present eye-tracking study, we tested 42 monolingual German-learning infants aged six to 14 months by means of a looking-while-listening paradigm. Infants saw two pictures side-by-side on a screen, whilst an unfamiliar male talker named one of both. Overall, infants did not fixate the target picture more than the distractor picture. In line with previous results, infants’ performance on the task was higher when the target and distractor word differed within their word frequency operationalized by the parental rating of word exposure. Therefore, our results further emphasize cross-linguistic differences in early word learning and strengthen the view that infants might use extra-linguistic cues within the stimulus pairing, such as frequency imbalance, to disambiguate between two potential referents.

Do Cross-Language Script Differences Enable Bilinguals to Function Selectively When Speaking in One Language Alone?

The present study examined the role of script in bilingual speech planning by comparing the performance of same and different-script bilinguals. Spanish-English bilinguals (Experiment 1) and Japanese-English bilinguals (Experiment 2) performed a picture-word interference task in which they were asked to name a picture of an object in English, their second language, while ignoring a visual distractor word in Spanish or Japanese, their first language. Results replicated the general pattern seen in previous bilingual picture-word interference studies for the same-script, Spanish-English bilinguals but not for the different-script, Japanese-English bilinguals. Both groups showed translation facilitation, whereas only Spanish-English bilinguals demonstrated semantic interference, phonological facilitation, and phono-translation facilitation. These results suggest that when the script of the language not in use is present in the task, bilinguals appear to exploit the perceptual difference as a language cue to direct lexical access to the intended language earlier in the process of speech planning.

Effects of orthography in the picture-word task: Evidence from Japanese scripts

The picture-word task presents participants with a number of pictured objects together with a written distractor word superimposed upon each picture, and their task is to name the depicted object while ignoring the distractor word. Depending on the specific picture and word combination, various effects, including the identity facilitation effect (e.g., DOG + dog) and the semantic interference effect (e.g., GOAT + cow), are often observed. The response patterns of the picture-word task in terms of naming latencies reflect the mechanisms underlying lexical selection in speech production. Research using this method, however, has typically focused on alphabetic languages, or involved bilingual populations, making it difficult to specifically investigate orthographic effects in isolation. In this paper, we report five experiments investigating the role of orthography in the picture-word task by varying distractor script (using the multiscriptal language Japanese, and pseudohomophonic spellings in English) across three different populations (Japanese monolinguals, Japanese-English bilinguals, and English monolinguals), investigating both the identity facilitation effect and the semantic interference effect. The results generally show that the magnitude of facilitation is affected by orthography even within a single language. The findings and specific patterns of results are discussed in relation to current theories on speech production.

Cross-linguistic interactions across modalities: Effects of the oral language on sign production

To investigate cross-linguistic interactions in bimodal bilingual production, behavioural and electrophysiological measures (ERPs) were recorded from 24 deaf bimodal bilinguals while naming pictures in Catalan Sign Language (LSC). Two tasks were employed, a picture-word interference and a picture-picture interference task. Cross-linguistic effects were explored via distractors that were either semantically related to the target picture, to the phonology/orthography of the Spanish name of the target picture, or were unrelated. No semantic effects were observed in sign latencies, but ERPs differed between semantically related and unrelated distractors. For the form-related manipulation, a facilitation effect was observed both behaviourally and at the ERP level. Importantly, these effects were not influenced by the type of distractor (word/picture) presented providing the first piece of evidence that deaf bimodal bilinguals are sensitive to oral language in sign production. Implications for models of cross-linguistic interactions in bimodal bilinguals are discussed.

Intersubject variability in spoken verb production: Effects of hierarchy and transitivity

In two experiments employing the picture-word interference (PWI) paradigm, we explored how a verb’s hierarchy and transitivity influences its retrieval during spoken production. Experiment 1 involved transitive (i.e. object-oriented, e.g. eat) action pictures accompanied by a to-be-ignored distractor word that was either a related coordinate (‘drink’) or troponym (‘devour’), while Experiment 2 employed intransitive (e.g. walk) stimuli. Assuming these relationships operate similarly for verbs and object nouns, we expected to observe faster naming times for troponyms, and slower naming times for coordinates. Conventional group-level analyses of the null average hypothesis revealed no significant effects in either experiment. However, analyses of the global null hypothesis revealed significant interindividual variability for troponym distractors in Experiment 1, with a similar trend in Experiment 2. These results indicate verbs may have a different conceptual-lexical organisation to object nouns in the mental lexicon, less constrained by hierarchical categories, with their processing more influenced by subject-specific variables.

Internet based language production research with overt articulation: Proof of concept, challenges, and practical advice

Language production experiments with overt articulation have thus far only scarcely been conducted online, mostly due to technical difficulties related to measuring voice onset latencies. Especially the poor audiovisual synchrony in web experiments (Bridges et al., 2020) is a challenge to time-locking stimuli and participants’ spoken responses. We tested the viability of conducting language production experiments with overt articulation in online settings using the picture-word interference paradigm – a classic task in language production research. In three pre-registered experiments (N=48 each), participants named object pictures while ignoring visually superimposed distractor words. We implemented a custom voice recording option in two different web experiment builders and recorded naming responses in audio files. From these stimulus-locked audio files we extracted voice onset latencies offline. In a control task, participants classified the last letter of a picture name as a vowel or consonant via button-press, a task that shows comparable semantic interference effects. We expected slower responses when picture and distractor word were semantically related compared to unrelated, independently of task. This semantic interference effect is robust, but relatively small. It should therefore crucially depend on precise timing. We replicated this effect in an online setting, both for button-press and overt naming responses, providing a proof of concept that naming latency - a key dependent variable in language production research - can be reliably measured in online experiments. We discuss challenges for online language production research and suggestions how to overcome them. The scripts for the online implementation are made available.

Contingency learning is not modulated by cognitive control demands

People form associations between stimuli and responses, resulting in faster responses to stimuli that occur more frequently. This type of stimulus-response contingency learning has often been claimed to confound putative effects of cognitive control in conflict tasks, like the Stroop task. However, the underlying assumption that contingency learning itself is not modulated by cognitive control demand remains unanswered. To assess whether the presence and congruency of distracters alters contingency learning, we had participants perform a face-gender Stroop classification task, either without (“nonconflict task”) or with congruent and incongruent distractor word labels overlaid (“conflict task”). Importantly, we also manipulated the frequency at which specific face images were displayed (1, 5, or 10 times) in order to measure contingency learning. If cognitive demand facilitated contingency learning, we would see more pronounced learning effects in the conflict task than in the nonconflict task. By contrast, if cognitive demand interfered with contingency learning, we would see less pronounced learning effects in the conflict task. Across a preregistered within-participant and a separate between-participant Experiment, we observed additive main effects for image frequency and task type, with a standard contingency learning effect, a standard Stroop congruency effect, and faster categorization of faces when no distractor word label was presented. Crucially, there was little evidence for an interaction effect between task and stimulus frequency, thus documenting that contingency learning occurs independently of cognitive demand in a typical conflict task. Implications for these findings are discussed.