Why the P600 is not just a P300: the role of the basal ganglia
Introduction
Starting with the first reports, a centro-parietal positive event-related potential (ERP) component (P600) has been associated with processing efforts related to syntactic aspects of language (Osterhout and Holcomb, 1992, Hagoort et al., 1993). Meanwhile, similar positive ERP deflections have been found for a broad range of syntactic anomalies, such as phrase structure violations (Friederici et al., 1999, Hahne and Friederici, 1999), morphosyntactic violations (Hagoort et al., 1993, Gunter et al., 1997, Coulson et al., 1998a), argument-structure and case violations (Friederici and Frisch, 2000, Frisch and Schlesewsky, 2001), or dispreferred continuations of temporarily ambiguous sentences (Osterhout and Holcomb, 1992, Mecklinger et al., 1995). As a consequence, the P600 ERP component is taken as a specific marker of syntactic aspects of language processing (as opposed to the lexical-semantic N400 effect) by many authors in the field (Osterhout and Holcomb, 1992, Friederici, 1995).
There is, however, an ongoing debate as to whether the P600 is indeed language specific or just a P300-like effect, indicating the detection of an unexpected, task-relevant target (Gunter et al., 1997, Coulson et al., 1998a). The classical experimental paradigm to elicit a P300 or, more precisely, the P3b subcomponent of the P300 complex (Picton, 1992) is the so-called auditory oddball paradigm. Subjects listen to a series of tones which can be classified into two different categories, with one of the two being task-relevant (i.e. deviant tones have to be counted) and evoking a P300 response. The fact that the P3b and the P600 have a very similar centro-parietal scalp distribution and that the amplitudes of both increase when the number of critical targets decreases (Donchin, 1981, Coulson et al., 1998a, Hahne and Friederici, 1999) has led some authors to believe that the two components are identical (Coulson et al., 1998a). Latency differences between the P600 and the P300 oddball effect are explained with the higher complexity of linguistic stimuli.
Opponents of this identity hypothesis argued that probability dependence holds for other components as well and that due to the inverse problem similar scalp distributions can in principle not reveal that both components rely upon the same neuronal generators (Osterhout and Hagoort, 1999). However, all opponents seem to agree that different P300 and P600 generators would support existence of two different components (Coulson et al., 1998b, Osterhout and Hagoort, 1999). This question, however, has not been settled yet.
As for the P3b, it is assumed that this component reflects activity from a number of different sources (Picton, 1992, Johnson, 1993, Mecklinger et al., 1998), which seem to be differentially involved as a function of the modality of the stimulus (Johnson, 1989, Rogers et al., 1991). For the auditory P3b, Rogers and colleagues described two main generators, one in the thalamic region and one in the posterior temporal lobe (Rogers et al., 1991). Studies using ERPs and functional magnetic resonance imaging (fMRI) suggest an involvement of the thalamus, the anterior portion of the gyrus cinguli as well as portions of the supramarginal gyrus (Menon et al., 1997). To our knowledge, however, the neural basis of the P600 component has not been specified so far.
In the present studies, we want to address the question of whether the neuronal sources of the two components are identical by testing two groups of aphasic patients (with versus without lesions involving the basal ganglia) in two experiments: one using grammatical violations and the second an auditory oddball.
Section snippets
Experiment 1
In the first experiment, subjects listened to grammatical and ungrammatical sentences. Ungrammatical sentences contained morphosyntactic violations. According to a number of previous studies, this violation elicits a P600 component in the ERP (Gunter et al., 1997, Coulson et al., 1998a).
Experiment 2
The results of Experiment 1 suggest that the observation of a P600 component in response to a grammatical violations depends on intact basal ganglia. The goal of Experiment 2 was to test whether the basal ganglia also play a crucial role in generating or mediating of the P3b component elicited by improbable, but task-relevant tones (auditory oddball paradigm).
Discussion
Taken together, the data from the two experiments shed clarifying light onto the question as to which brain structures are involved in the elicitation of the P300 and the P600 as well as onto the relationship between the two components.
Two groups of patients, one with a lesion involving the basal ganglia as well as one group in which this brain area was intact, were tested in an auditory ERP experiment with morphosyntactic violations as well as in an auditory oddball paradigm. As could be
Acknowledgements
We want to thank Anja Hahne, Korinna Eckstein, Juliane Hofmann and Angelika Wolf for their assistance in preparing the sentence materials, Claudia Misch for speaking the sentences and Ina Koch for recording the ERP data.
References (22)
The time course of syntactic activation during language processing: a model based on neuropsychological and neurophysiological data
Brain Lang
(1995)- et al.
Verb-argument structure processing: the role of verb-specific and argument-specific information
J Mem Lang
(2000) - et al.
A MEG analysis of the P300 in visual discrimination tasks
Electroenceph clin Neurophysiol
(1998) - et al.
Localization of the P3 sources in using magnetoencephalography and magnetic resonance imaging
Electroenceph clin Neurophysiol
(1991) - et al.
The CELEX lexical database
(1993) - et al.
Event-related brain potentials: an introduction
- et al.
Expect the unexpected: event-related brain response to morphosyntactic violations
Lang Cogn Proc
(1998) - et al.
ERPs and domain specificity: beating a straw horse
Lang Cogn Proc
(1998) Surprise … surprise?
Psychophysiology
(1981)- et al.
Lexical integration: sequential effects of syntactic and semantic information
Mem Cognit
(1999)
Selective vulnerability of early and late syntactic processes in Parkinson patients: ERP evidence
Brain Lang
Cited by (109)
Neural processes underlying statistical learning for speech segmentation in dogs
2021, Current BiologyThe electrophysiology of aphasia: A scoping review
2021, Clinical NeurophysiologyCitation Excerpt :None of the studies investigating therapy intervention examined the P600. Three studies (Friederici et al., 1999; Frisch et al. 2003; Kotz et al. 2003) took a lesion approach to examining ERP in people with aphasia (Table 2d). All three articles were produced by the same research team.
Hierarchical control as a shared neurocognitive mechanism for language and music
2021, CognitionCitation Excerpt :There is mounting evidence for a crucial role of the basal ganglia in linguistic syntax, particularly again in non-automatic, controlled syntactic processing (for reviews, see Jeon & Friederici, 2015; Kotz, Schwartze, & Schmidt-Kassow, 2009). For example, focal basal ganglia patients show abnormality in controlled syntactic processing (Frisch, Kotz, von Cramon, & Friederici, 2003; Kotz, Frisch, von Cramon, & Friederici, 2003). In a similar vein, both PD and HD patients show impairments in handling sentences with non-canonical syntactic complexity, ambiguity, or embedding (Birba et al., 2017), while they do not show impairments in automatic syntactic processing (Friederici, Kotz, et al., 2003; Longworth, 2005; Teichmann, Dupoux, Kouider, & Bachoud-Lévi, 2006).