On the role of bias in dissociated phonological priming effects: A reply to Goldinger (1999).

Phonological priming studies have revealed two dissociated effects: low-similarity facilitation and highsimilarity interference (Hamburger & Slowiaczek, 1996; Slowiaczek & Hamburger, 1992). Because these two effects are influenced by different variables, they most likely reflect different processes that occur during auditory word recognition. Goldinger (1999) suggests that one bias is responsible for all phonological priming effects. In this reply, we argue against such a position. Although low similarity facilitation is likely the product of this bias, the data on phonological priming indicate that the dissociated highsimilarity interference cannot be produced by the same mechanism. Instead, the data indicate that high-similarity interference may reflect lexical processes.

Repetition priming and experimental context effects.

The repetition priming effect has been taken as evidence that lexical entries are automatically activated in memory. However, contextual cues in an experiment may influence repetition priming, suggesting that this priming may not be the result of automatic processes. Five experiments used a primed shadowing task to explore the role of phonological context on repetition priming. In the first four experiments, high- and low-expectancy conditions were tested. Between experiments, the degree of phonological relatedness (high or low phonological similarity), percentage of related trials that were repetition trials (ID ratio), and modality of stimulus presentation (auditory or visual) were manipulated. Results indicate that repetition priming in auditory word recognition is eliminated by a low ID ratio. The fact that an external variable such as the ID ratio influences repetition priming suggests that repetition priming may be the product of processes extraneous to the automatic activation of lexical entries.

Phonological priming reflects lexical competition.

A phonological relationship between a prime and a target produces facilitation when one or two initial phonemes are shared (low-similarity facilitation) but produces interference when more phonemes are shared (high-similarity interference; Slowiaczek & Hamburger, 1992). Although low-similarity facilitation appears to be a strategic effect (Goldinger, Luce, Pisoni, & Marcario, 1992), this result cannot generalize to high-similarity interference because the two effects are dissociated (Slowiaczek & Hamburger, 1992). In the present study, strategic processing in high-similarity interference was investigated. The phonological relatedness proportion (PRP) and the prime-target interstimulus interval (ISI) were varied in a shadowing experiment. Low-similarity facilitation was found only with a high PRP and long ISI, but high-similarity interference was found regardless of PRP and ISI. These results suggest that strategies influence low-similarity facilitation, but high-similarity interference reflects automatic processing.

Prelexical facilitation and lexical interference in auditory word recognition.

Phonological priming effects were examined in an auditory single-word shadowing task. In 6 experiments, target items were preceded by auditorily or visually presented, phonologically similar, word or nonword primes. Results revealed facilitation in response time when a target was preceded by a word or nonword prime having the same initial phoneme when the prime was auditorily presented but not when it was visually presented. Second, modality-independent interference was observed when the phonological overlap between the prime and target increased from 1 to 3 phonemes for word primes but not for nonword primes. Taken together, these studies suggest that phonological information facilitates word recognition as a result of excitation at a prelexical level and increases response time as a result of competition at a lexical level. These processes are best characterized by connectionist models of word recognition.

Information selection and use in hypothesis testing: what is a good question, and what is a good answer?

The process of hypothesis testing entails both information selection (asking questions) and information use (drawing inferences from the answers to those questions). We demonstrate that although subjects may be sensitive to diagnosticity in choosing which questions to ask, they are insufficiently sensitive to the fact that different answers to the same question can have very different diagnosticities. This can lead subjects to overestimate or underestimate the information in the answers they receive. This phenomenon is demonstrated in two experiments using different kinds of inferences (category membership of individuals and composition of sampled populations). In combination with certain information-gathering tendencies, demonstrated in a third experiment, insensitivity to answer diagnosticity can contribute to a tendency toward preservation of the initial hypothesis. Results such as these illustrate the importance of viewing hypothesis-testing behavior as an interactive, multistage process that includes selecting questions, interpreting data, and drawing inferences.

Stress and context in auditory word recognition.

Recent work in psycholinguistics has revealed that the role of lexical stress in auditory word recognition may be a complex one involving other potential sources of information. In the present investigation, the nature of lexical stress effects on auditory word recognition in context is examined. The ability of subjects to identify words based on the prosodic pattern of the word is examined for the words in isolation and in a sentence context. The results revealed a small influence of stress on the identification of the stimulus waveforms relative to a large effect of context. The data indicate that lexical stress is used in the identification of the words; however, its role is minor compared to the importance of contextual information.

Effects of lexical stress in auditory word recognition.

Although research examining the use of prosodic information in the processing of spoken words has increased in recent years, results from these studies have been inconclusive. The present series of experiments systematically examines the importance of one prosodic variable (lexical stress) in the recognition of isolated spoken words. Data collected in an identification task suggest that segmental information may be more heavily relied upon when appropriate lexical stress information is not available. Results of subsequent reaction time experiments support the hypothesis that lexical stress influences the processing of auditorily presented words. In three shadowing experiments, correctly stressed items were produced faster than incorrectly stressed items, and in a lexical decision experiment, correctly stressed words were classified faster than incorrectly stressed words. Thus, this work provides evidence across several experimental tasks for the use of lexical stress information in the processing of spoken words. Moreover, the data suggest that lexical stress should be an important aspect of the representation of words in an interactive model of auditory word recognition.

Phonological priming in auditory word recognition.

Cohort theory, developed by Marslen-Wilson and Welsh (1978), proposes that a "cohort" of all the words beginning with a particular sound sequence will be activated during the initial stage of the word recognition process. We used a priming technique to test specific predictions regarding cohort activation in three experiments. In each experiment, subjects identified target words embedded in noise at different signal-to-noise ratios. The target words were either presented in isolation or preceded by a prime item that shared phonological information with the target. In Experiment 1, primes and targets were English words that shared zero, one, two, three, or all phonemes from the beginning of the word. In Experiment 2, nonword primes preceded word targets and shared initial phonemes. In Experiment 3, word primes and word targets shared phonemes from the end of a word. Evidence of reliable phonological priming was observed in all three experiments. The results of the first two experiments support the assumption of activation of lexical candidates based on word-initial information, as proposed in cohort theory. However, the results of the third experiment, which showed increased probability of correctly identifying targets that shared phonemes from the end of words, did not support the predictions derived from the theory. The findings are discussed in terms of current models of auditory word recognition and recent approaches to spoken-language understanding.