Speech Monitoring Research Articles

The Sensory Consequences of Speaking: Parametric Neural Cancellation during Speech in Auditory Cortex

When we speak, we provide ourselves with auditory speech input. Efficient monitoring of speech is often hypothesized to depend on matching the predicted sensory consequences from internal motor commands (forward model) with actual sensory feedback. In this paper we tested the forward model hypothesis using functional Magnetic Resonance Imaging. We administered an overt picture naming task in which we parametrically reduced the quality of verbal feedback by noise masking. Presentation of the same auditory input in the absence of overt speech served as listening control condition. Our results suggest that a match between predicted and actual sensory feedback results in inhibition of cancellation of auditory activity because speaking with normal unmasked feedback reduced activity in the auditory cortex compared to listening control conditions. Moreover, during self-generated speech, activation in auditory cortex increased as the feedback quality of the self-generated speech decreased. We conclude that during speaking early auditory cortex is involved in matching external signals with an internally generated model or prediction of sensory consequences, the locus of which may reside in auditory or higher order brain areas. Matching at early auditory cortex may provide a very sensitive monitoring mechanism that highlights speech production errors at very early levels of processing and may efficiently determine the self-agency of speech input.

Investigating the inner speech of people who stutter: Evidence for (and against) the Covert Repair Hypothesis

In their Covert Repair Hypothesis, Postma and Kolk (1993) suggest that people who stutter make greater numbers of phonological encoding errors, which are detected during the monitoring of inner speech and repaired, with stuttering-like disfluencies as a consequence. Here, we report an experiment that documents the frequency with which such errors are made. Thirty-two people who stutter (PWS) and thirty-two normally fluent controls, matched for age, gender and education, recited tonguetwisters and self-reported any errors they perceived themselves to have made. In 50% of trials the tonguetwisters were recited silently and errors reported were those detected in inner speech. Compared to controls, PWS produced significantly more word-onset and word-order errors. Crucially, this difference was found in inner as well as in overt speech. Comparison of experimenter ratings and participants’ own self-ratings of their overt speech revealed similar levels of accuracy across the two groups, ruling out a suggestion that PWS were simply more sensitive to the errors they made. However, the frequency of participants’ inner-speech errors was not correlated to their SSI4 scores, nor to two other measures of stuttering severity. Our findings support Postma and Kolk's contention that, when speech rate is held constant, PWS make, and therefore detect, more errors of phonological encoding. They do not, however, support the hypothesis that stuttering-like disfluencies in everyday speech stem from covert repairs of errors of phonological encoding.Learning outcomes: Readers will learn about three current psycholinguistic theories of stuttering, and how speech-errors elicited during tonguetwister recitation can be used to explore the controversies that exist surrounding: (a) Whether or not people who stutter are more prone to making language production errors; and (b) The extent to which stuttering-like disfluencies stem from covert repairs of language-production errors.

Investigation of self-monitoring in fluent aphasia with jargon

Background: Some individuals with fluent aphasia produce jargon errors in speech production. Jargon likely results from derailed encoding operations required for language production, although the exact mechanism remains debated. It is also unclear if persons who produce jargon are able to self-monitor; that is, detect the non-word status of their imminent utterance and self-correct the error. Aims: This study investigated the characteristics of speech monitoring in individuals with fluent aphasia who produce jargon. The association of speech monitoring with amount of jargon produced and success in auditory comprehension was examined. Further, the role of post-articulatory auditory feedback on speech-monitoring success was examined by introducing an auditory masker (white noise). Finally, the influence of task demands on amount of jargon and speech monitoring was examined. Methods & Procedures: Five individuals with fluent aphasia with jargon participated in the study. Speech monitoring was tested using three production tasks (picture naming, non-word repetition, and word repetition) under two listening conditions (normal listening and masking noise). Speech-monitoring score was derived from self-judgements of production accuracy. Auditory processing was examined in speech discrimination, lexical decision, and single-word identification tasks. Outcomes & Results: Results indicated that severity of self-monitoring impairment is: (1) strongly correlated with severity of jargon production, (2) poorly correlated with auditory processing abilities, (3) worse under masking noise, and (4) worse for naming compared to repetition. Conclusions: Overall, all five participants with jargon demonstrated severely impaired speech monitoring under both listening conditions across all three production tasks, suggesting that the role of speech-monitoring impairment in jargon cannot be disregarded. The increase in amount of jargon and severity of speech-monitoring failure with increased linguistic demands of the production task (as in picture naming and non-word repetition) raises the possibility of a post-semantic deficit in accessing phonological codes. This possible locus of deficit, and implications for understanding production of jargon, are discussed.

Effective connectivity of fMRI data using ancestral graph theory: Dealing with missing regions

Most of the current methods to assess effective connectivity from functional magnetic resonance imaging (fMRI) rely on the assumption that all relevant brain regions are entered into the analysis. If this assumption is untenable, which we believe is most often the case, then spurious connections between brain regions can appear. In this paper we propose to use an ancestral graph to model connectivity, which provides a way to avoid spurious connections. The ancestral graph is determined from trial-by-trial variation and not from the time series. A random effects model is defined for ancestral graphs which allows for individual differences in terms of graph parameters (e.g., connection strength). Procedures for model selection, model fit, and hypothesis testing of ancestral graphs are proposed. The hypothesis test can be used to find differences in connection strength between, for example, conditions. Monte Carlo simulations show that the ancestral graph is appropriate to model connectivity from fMRI condition specific trial data. To assess the accuracy further, the proposed method is applied to real fMRI data to determine how brain regions interact during speech monitoring.

Rhythmic Versus Phonemic Interference in Delayed Auditory Feedback

Delayed auditory feedback (DAF) of a speaker's voice disturbs normal speech production. Various traditional theories assume that the content of the delayed feedback signal interferes with the actual production of a particular speech unit (phonemic content hypothesis). The displaced rhythm hypothesis as an alternative explanation suggests that speech disturbances arise from a disruptive rhythm that is produced by the delayed speech signal. The present experimental study directly contrasted the role of rhythm and speech content in a DAF task using speech units as stimuli. One hundred fifty-one participants read aloud 4 different sequences of double syllables that varied in phonemic content and rhythm while auditory feedback was either nondelayed or delayed by 200 or 400 ms. In line with previous studies, the authors found a peak of disturbances at a delay of about 200 ms, independent of speech rate. More important, the present results clearly support the displaced rhythm hypothesis. A speech rate dependency of this effect was also found. Rhythm seems to be a significant criterion of speech monitoring, and hence a mismatch between spoken words and auditory feedback realized by DAF induces obvious speech problems on rhythmic level regardless of phonemic discrepancy at the same time.

Monitoring of self‐generated speech in adolescents with 22q11.2 deletion syndrome

The present report examines the monitoring of self-generated speech in adolescents with 22q11.2 deletion syndrome (22q11DS), a neurogenetic disorder associated with very high risk for psychosis. Between-participant group design. In this study, 20 adolescents with 22q11DS, 19 age- and IQ-matched controls, and 19 typically developing adolescents were enrolled. Participants completed a speech-monitoring task, in which they were asked to silently or overtly read a series of word and non-word items. Subjects then filled out a recognition sheet containing studied and novel items. They were asked to identify the previously studied item, and to attribute the reading condition (silent vs. overt) under which each recognized item was encoded. Adolescents with 22q11DS commit more external attribution errors compared to both control groups, by exhibiting an increased tendency to report silently read items as though they had been read overtly. Further, results suggest that increased cognitive effort exacerbates the external attribution tendency in adolescents with 22q11DS. Increased internal attributions were also observed in the IQcontrol and 22q11DS groups in comparison to typically developing adolescents. Similarly to adult individuals exhibiting positive symptoms of psychosis, adolescents with 22q11DS exhibit an external attribution bias for inner speech. This bias seems to be exacerbated by increased cognitive effort, suggesting a failure to recollect information pertaining to cognitive operations during self-monitoring. Cognitive biases associated to schizophrenia may be detected in adolescents at very high risk for psychosis. These observations provide further evidence for the presence of an external attribution bias along the clinical continuum of psychosis vulnerability.

Design of an intelligent system for speech monitoring and treatment of low and excessive vocal intensity

A number of people, and especially children, suffer from speech problems such as vocal intensity that is too low or is often excessive, and this is one of the behavioral problems that may be encountered in people with an intellectual disability. It may also occur alongside aggressive behavior which produces disturbances, and which is objected to by others. People with this behavior may often be unaware of their problem. To limit these negative outcomes, an intelligent system has been designed to monitor speech intensity outside the clinic, to provide the patient with information, and to control the treatment of excessive or low vocal intensity without the need for continuous therapeutic follow-up in a speech pathology clinic. The use of this intelligent system proves to be effective because of its many features. The technical design considerations, the system features, and a system evaluation will be discussed.

Functional Magnetic Resonance Imaging of Inner Speech in Schizophrenia

Auditory verbal hallucinations in schizophrenia have been linked to defective monitoring of one's own verbal thoughts. Previous studies have shown that patients with auditory verbal hallucinations show attenuated activation of brain regions involved with auditory processing during the monitoring of inner speech. However, there are no functional magnetic resonance imaging studies explicitly comparing the perception of external speech with internal speech in the same patients with schizophrenia. The present study investigated the functional neuroanatomy of inner and external speech in both patients with schizophrenia and healthy control subjects. Fifteen patients with schizophrenia and 12 healthy control subjects were studied using functional magnetic resonance imaging while listening to sentences or imagining sentences. Significant interactions between group (control subjects vs. patients) and task (listening vs. inner speech) were seen for the left superior temporal gyrus, as well as regions within the cingulate gyrus. Attenuated deactivation of the left superior temporal gyrus in schizophrenia patients during the processing of inner speech may reflect deficits in the forward models subserving self-monitoring.

Neural network of speech monitoring overlaps with overt speech production and comprehension networks: A sequential spatial and temporal ICA study

The neural correlates of speech monitoring overlap with neural correlates of speech comprehension and production. However, it is unclear how these correlates are organized within functional connectivity networks, and how these networks interact to subserve speech monitoring. We applied spatial and temporal independent component analysis (sICA and tICA) to a functional magnetic resonance imaging (fMRI) experiment involving overt speech production, comprehension and monitoring. SICA and tICA respectively decompose fMRI data into spatial and temporal components that can be interpreted as distributed estimates of functional connectivity and concurrent temporal dynamics in one or more regions of fMRI activity. Using sICA we found multiple connectivity components that were associated with speech perception (auditory and left fronto-temporal components) and production (bilateral central sulcus and default-mode components), but not with speech monitoring. In order to further investigate if speech monitoring could be mapped in the auditory cortex as a unique temporal process, we applied tICA to voxels of the sICA auditory component. Amongst the temporal components we found a single, unique component that matched the speech monitoring temporal pattern. We used this temporal component as a new predictor for whole-brain activity and found that it correlated positively with bilateral auditory cortex, and negatively with the supplementary motor area (SMA). Psychophysiological interaction analysis of task and activity in bilateral auditory cortex and SMA showed that functional connectivity changed with task conditions. These results suggest that speech monitoring entails a dynamic coupling between different functional networks. Furthermore, we demonstrate that overt speech comprises multiple networks that are associated with specific speech-related processes. We conclude that the sequential combination of sICA and tICA is a powerful approach for the analysis of complex, overt speech tasks.

Hallucinations and two types of free-recall intrusion in schizophrenia

Previous research has demonstrated that various types of verbal source memory error are associated with positive symptoms in patients with schizophrenia. Notably, intrusions in free recall have been associated with hallucinations and delusions. We tested the hypothesis that extra-list intrusions, assumed to arise from poor monitoring of internally generated words, are associated with verbal hallucinations and that intra-list intrusions are associated with global hallucination scores. A sample of 41 patients with schizophrenia was administered four lists of words, followed by free recall. The number of correctly recalled words and the number of extra- and intra-list intrusions were tallied. The verbal hallucination score was significantly correlated with the number of extra-list intrusions, whereas it was unrelated to the number of correctly recalled words. The number of intra-list intrusions was significantly correlated with the global, but not with the verbal, hallucination score in the subsample of hallucinating patients. It was marginally significantly correlated with the delusion score in delusional patients. The data corroborate the view that verbal hallucinations are linked to defective monitoring of internal speech, and that errors in context processing are involved in hallucinations and delusions.

Functional MRI of Verbal Self-monitoring in Schizophrenia: Performance and Illness-Specific Effects

Previous small-sample studies have shown altered frontotemporal activity in schizophrenia patients with auditory hallucinations and impaired monitoring of self-generated speech. We examined a large cohort of patients with schizophrenia (n = 63) and a representative group of healthy controls (n = 20) to disentangle performance, illness, and symptom-related effects in functional magnetic resonance imaging–detected brain abnormalities during monitoring of self- and externally generated speech in schizophrenia. Our results revealed activation of the thalamus (medial geniculate nucleus, MGN) and frontotemporal regions with accurate monitoring across all participants. Less activation of the thalamus (MGN, pulvinar) and superior-middle temporal and inferior frontal gyri occurred in poorly performing patients (1 standard deviation below controls’ mean; n = 36), relative to the combined group of controls and well-performing patients. In patients, (1) greater deactivation of the ventral striatum and hypothalamus to own voice, combined with nonsignificant activation of the same regions to others’ voice, associated positively with negative symptoms (blunted affect, emotional withdrawal, poor rapport, passive social avoidance) regardless of performance and (2) exaggerated activation of the right superior-middle temporal gyrus during undistorted, relative to distorted, feedback associated with both positive symptoms (hallucinations, persecution) and poor performance. A further thalamic abnormality characterized schizophrenia patients regardless of performance and symptoms. We conclude that hypoactivation of a neural network comprised of the thalamus and frontotemporal regions underlies impaired speech monitoring in schizophrenia. Positive symptoms and poor monitoring share a common activation abnormality in the right superior temporal gyrus during processing of degraded speech. Altered striatal and hypothalamic modulation to own and others’ voice characterizes emotionally withdrawn and socially avoidant patients.

Intervention for a Child with Auditory Neuropathy/Dys-synchrony

Intervention for a Child with Auditory Neuropathy/Dys-synchrony

Uncontrollable voices and their relationship to gating deficits in schizophrenia

BackgroundPrepulse inhibition (PPI) of the startle response refers to the ability of a weak prestimulus to transiently inhibit the response to a closely following strong sensory stimulus. This effect is reduced in a number of disorders known to be associated with impaired gating of sensory, cognitive or motor information. The aim of this study was to investigate PPI deficit in relation to the dimensions of auditory hallucinations in patients with schizophrenia or schizoaffective disorder. MethodPPI of the acoustically elicited eye blink startle response was measured electromyographically in 62 patients with schizophrenia (n=55) or schizoaffective disorder (n=7) (26 of 62 with current auditory hallucinations) and 22 healthy participants matched, on average, to age and sex of the patient group. ResultsPatients, as a group, showed reduced PPI compared to healthy participants. The presence of auditory hallucinations was associated with a marked PPI deficit if the patients felt that they had no control over their occurrence and that they were unable to dismiss them. Hearing voices with a high degree of negative content was associated with high mean startle amplitude in patients with current auditory hallucinations. ConclusionsAlthough auditory hallucinations in patients with schizophrenia are theorised to result from impaired monitoring of inner speech, the inability to consciously ignore them appears to be associated with a gating deficit. Hearing voices with negative content is associated with hyper-startle responding, possibly because such voices are threatening and thus provoke anxiety.

Capacity demands of phoneme selection in word production: New evidence from dual-task experiments.

Three dual-task experiments investigated the capacity demands of phoneme selection in picture naming. On each trial, participants named a target picture (Task 1) and carried out a tone discrimination task (Task 2). To vary the time required for phoneme selection, the authors combined the targets with phonologically related or unrelated distractor pictures (Experiment 1) or words, which were clearly visible (Experiment 2) or masked (Experiment 3). When pictures or masked words were presented, the tone discrimination and picture naming latencies were shorter in the related condition than in the unrelated condition, which indicates that phoneme selection requires central processing capacity. However, when the distractor words were clearly visible, the facilitatory effect was confined to the picture naming latencies. This pattern arose because the visible related distractor words facilitated phoneme selection but slowed down speech monitoring processes that had to be completed before the response to the tone could be selected.

SELF-MONITORING AND SELF-REPAIR IN SPONTANEOUS SPEECH

This study explores what repairs in the spontaneous production of speech reveal about the psycholinguistic processes of self-monitoring and self-repair. Three intervals were examined: error-to-cut off; cut off-to-repair; error-to-repair. The intervals indicate support theories of internal speech monitoring, and also indicate that the planning of speech-repairs can take place pre-articulatorily as well

Neural correlates of verbal feedback processing: An fMRI study employing overt speech

Speakers use external auditory feedback to monitor their own speech. Feedback distortion has been found to increase activity in the superior temporal areas. Using fMRI, the present study investigates the neural correlates of processing verbal feedback without distortion. In a blocked design, the following conditions were presented: (1) overt picture-naming, (2) overt picture-naming while pink noise was presented to mask external feedback, (3) covert picture-naming, (4) listening to the picture names (previously recorded from participants' own voices), and (5) listening to pink noise. The results show that auditory feedback processing involves a network of different areas related to general performance monitoring and speech-motor control. These include the cingulate cortex and the bilateral insula, supplementary motor area, bilateral motor areas, cerebellum, thalamus and basal ganglia. Our findings suggest that the anterior cingulate cortex, which is often implicated in error-processing and conflict-monitoring, is also engaged in ongoing speech monitoring. Furthermore, in the superior temporal gyrus, we found a reduced response to speaking under normal feedback conditions. This finding is interpreted in the framework of a forward model according to which, during speech production, the sensory consequence of the speech-motor act is predicted to attenuate the sensitivity of the auditory cortex.

Inner speech models of auditory verbal hallucinations: Evidence from behavioural and neuroimaging studies

A range of psychological theories have been proposed to account for the experience of auditory hallucinations in patients with psychosis. Influential amongst these theories are those implicating the defective monitoring of inner speech. Furthermore, self-monitoring and inner speech models have been the most studied using functional imaging. The aim of this article is to review the behavioural and neuroimaging evidence for the impaired monitoring of inner speech in patients who experience auditory verbal hallucinations. A comprehensive literature search was conducted for research investigating inner speech and cognitive self-monitoring models of auditory hallucinations in patients with schizophrenia. The evidence is critically discussed and directions for future investigations are suggested.

Perceptual uniqueness point effects in monitoring internal speech

Disagreement exists about how speakers monitor their internal speech. Production-based accounts assume that self-monitoring mechanisms exist within the production system, whereas comprehension-based accounts assume that monitoring is achieved through the speech comprehension system. Comprehension-based accounts predict perception-specific effects, like the perceptual uniqueness-point effect, in the monitoring of internal speech. We ran an extensive experiment testing this prediction using internal phoneme monitoring and picture naming tasks. Our results show an effect of the perceptual uniqueness point of a word in internal phoneme monitoring in the absence of such an effect in picture naming. These results support comprehension-based accounts of the monitoring of internal speech.

Halting in single word production: A test of the perceptual loop theory of speech monitoring

The perceptual loop theory of speech monitoring ( Levelt, 1983) claims that inner and overt speech are monitored by the comprehension system, which detects errors by comparing the comprehension of formulated utterances to originally intended utterances. To test the perceptual loop monitor, speakers named pictures and sometimes attempted to halt speech in response to auditory (Experiments 1 and 3) or visual (Experiments 2, 4, and 5) words that differed from the picture name. These stop-signal words were varied in terms of their semantic or phonological similarity to the intended word. The ability to halt word production was sensitive to phonological similarity and, in Experiment 5, to emotional valence, but not to semantic similarity. These results suggest that the perceptual loop detects errors by making comparisons at a level where phonological knowledge is represented. These data also imply that dialogue, back channeling, and other areas where speech production is affected by simultaneous comprehension may operate based on phonological comparisons.

Lingual kinematic strategies used to increase speech rate: Comparison between younger and older adults

The primary objective of this study was to assess the lingual kinematic strategies used by younger and older adults to increase rate of speech. It was hypothesised that the strategies used by the older adults would differ from the young adults either as a direct result of, or in response to a need to compensate for, age‐related changes in the tongue. Electromagnetic articulography was used to examine the tongue movements of eight young (M = 26.7 years) and eight older (M = 67.1 years) females during repetitions of /ta/ and /ka/ at a controlled moderate rate and then as fast as possible. The younger and older adults were found to significantly reduce consonant durations and increase syllable repetition rate by similar proportions. To achieve these reduced durations both groups appeared to use the same strategy, that of reducing the distances travelled by the tongue. Further comparisons at each rate, however, suggested a speed‐accuracy trade‐off and increased speech monitoring in the older adults. The results may assist in differentiating articulatory changes associated with normal aging from pathological changes found in disorders that affect the older population.