Finding meaning in novel geometric shapes influences electrophysiological correlates of repetition and dissociates perceptual and conceptual priming
Introduction
A stimulus is perceived. Later, it is perceived again. What differs between these two events? One way to address this question is to identify changes in how subjects and neurons respond to stimuli that repeat during memory tests. Priming refers to a change in the speed, accuracy, or other aspect of a behavioral response to a stimulus based on prior exposure to the same stimulus or to a related stimulus (Schacter, 1987, Richardson-Klavehn and Bjork, 1988, Roediger, 1990). Priming tests are implicit measures of memory in that subjects are not required to indicate overtly that stimuli are repeating, as they would in a recall or recognition test, though they may incidentally realize this (Richardson-Klavehn and Bjork, 1988).
Perceptual priming and conceptual priming are behaviorally distinct expressions of priming defined based on the nature of the information processing steps responsible for the repetition effects. Priming for repeated, physical features of stimuli is considered to underlie perceptual priming, whereas priming for stimulus meaning, independent from physical properties, underlies conceptual priming. Stimuli such as words and nameable pictures can engender both types of priming: perceptual priming for the visual word form and conceptual priming for word meaning, for instance. Changing the physical form of a stimulus from one presentation to the next (e.g., a word first read, then heard) should preferentially reduce perceptual priming, and manipulations such as format-switching are frequently used to dissociate behavioral and neural correlates of perceptual and conceptual priming (Henson, 2003, Schacter et al., 2004).
Perceptual and conceptual priming have been associated with changes in neural processing at distinct loci of the ventral visual-processing stream (Henson, 2003, Schacter et al., 2007), with conceptual priming involving neural repetition effects that are more anterior than those for perceptual priming. However, both forms of priming have been studied primarily using categories of well-learned and conceptually rich stimuli, such as words and nameable pictures. Indeed, some investigators have failed to find neural repetition effects for stimuli without pre-experimental familiarity (Rugg and Doyle, 1994, Crites et al., 2000, Schendan and Maher, 2009). Here we sought to determine the extent to which perceptual and conceptual priming can occur for novel geometric shapes and whether these different memory expressions occur in conjunction with characteristic neural repetition effects.
We measured event-related brain potential (ERP) correlates of perceptual and conceptual priming to obtain millisecond-by-millisecond observations of the neural activity associated with each memory expression. This level of temporal resolution is important because perceptual processing is thought to precede conceptual processing; perceptual processing steps (as in contour detection, figure-ground segregation, and perceptual grouping) are considered prerequisites for object categorization and retrieval of associated conceptual information (Biederman, 1987, Schendan and Kutas, 2002), although pre-experimental knowledge about meaningful stimuli can also influence perceptual processing (e.g., Peterson and Enns, 2005). The notion that perceptual and conceptual priming are distinct neural processes that potentially operate on distinct memory representations is generally consistent with theories of multiple memory systems (Tulving, 1972, Tulving and Schacter, 1990, Squire and Zola-Morgan, 1991), whereas some grounded-cognition theories of knowledge (Barsalou, 2008) would allow for similar neural operations underlying perceptual and conceptual processing.
We measured perceptual and conceptual processing using two implicit memory tests. Perceptual priming was measured during a perceptual task, loop detection, and conceptual priming was measured during a conceptual task involving meaning ratings. For both tasks, the study phase involved a meaning rating task. This design provided behavioral correlates of perceptual and conceptual priming that allowed their ERP correlates to be compared. Furthermore, we used minimalist geometric shapes, which we refer to as “squiggles,” to gain leverage on dissociating perceptual and conceptual processing, as described below.
Based on previous results obtained using the same stimulus set (Voss and Paller, 2007), we predicted that conceptual priming would occur selectively for the shapes given the highest meaningfulness ratings, and that conceptual priming of meaningful shapes would occur in conjunction with repetition effects on FN400 brain potentials. The FN400 is a negative ERP deflection between approximately 300 and 500 ms that is largest at frontocentral scalp locations and is reduced (i.e., more positive) for repeated relative to new items. In our previous study, we found evidence that conceptual priming occurred only for the most meaningful squiggles, although ERPs were recorded during a recognition test, not during an implicit memory test. FN400 potentials during recognition were reduced for meaningful squiggles, which were also able to support conceptual priming; squiggles that carried less meaning did not appear to support conceptual priming and did not exhibit FN400 repetition effects during the recognition test. These predictions are at odds with the dominant interpretation of FN400 potentials as generic markers of familiarity during episodic memory tests (reviewed in Rugg and Curran, 2007), but are consistent with recent results linking FN400 potentials to conceptual processing (Voss and Paller, 2006, Voss and Paller, 2007, Voss and Paller, 2009, Danker et al., 2008, Voss et al., in press).
Perceptual priming of visual shapes, typically nameable objects, has been associated with ERP repetition effects that onset earlier than FN400 potentials and index posterior cortical processing associated with visual perception (e.g., Allison et al., 1999). For example, frontocentral P150 potentials are larger for repeated than new objects between 120 and 200 ms (Schendan and Kutas, 2003) and are thought to index perceptual categorization processes (Schendan et al., 1998). Furthermore, lateral occipitotemporal P200 potentials between 190 and 270 ms are larger for new than repeated fragmented line drawings and are thought to index perceptual grouping processes (Schendan and Kutas, 2007a). We therefore predicted that ERP perceptual priming effects would occur earlier than FN400 effects associated with conceptual priming, which typically occur between 300 and 500 ms. However, neural correlates of perceptual priming might differ for squiggles versus highly familiar nameable objects.
We also predicted that incidental retrieval of episodic information from the study phase would occur during the priming tests despite the fact that the tests were indirect (Richardson-Klavehn and Bjork, 1988). Furthermore, this incidental retrieval would be associated with late-onset positive postentials with a posterior distribution, termed late positive complex or LPC potentials (Friedman and Johnson, 2000, Rugg and Curran, 2007, Voss and Paller, 2008). Indeed, explicit memory for study-phase episodes for squiggle stimuli is highly correlated with LPC potentials (Voss and Paller, 2007).
Section snippets
Subjects
Behavioral and electrophysiological data were collected from 15 Northwestern University students after informed consent was obtained. Five subjects were male, and all were right-handed, native English speakers between 18 and 24 years of age with normal or corrected-to-normal vision.
Materials
Visual stimuli were 300 minimalist visual shapes known as “squiggles” (Fig. 1). Squiggles were created by random hand-deformation of a square, circle, or triangle (Groh-Bordin et al., 2006), and were presented on a
Encoding
Results were collapsed across the 10 blocks for analyses because at the time of encoding subjects were unaware of the subsequent test format. The proportion of squiggles garnering meaningfulness ratings of high, medium, low, and none was 0.17 (SE = 0.02), 0.27 (SE = 0.01), 0.33 (SE = 0.02), and 0.23 (SE = 0.02), respectively. The high-meaning category (ratings high and medium together) comprised 44% (SE = 3%) of squiggles and the low-meaning category (ratings low and none together) comprised 56% (SE = 3%)
Discussion
Memory for squiggles, as assessed in tests of perceptual priming and conceptual priming, varied systematically with whether stimuli were meaningful. Ratings of meaningfulness given to each squiggle diverged across individuals based on the idiosyncratic extent to which conceptual knowledge in long-term memory was activated. Accessing such conceptual knowledge involves deciding that the stimulus is known based on pre-experimental experiences, akin to when an object's category is determined via
Acknowledgments
Financial support was provided by the United States National Science Foundation under grants 0518800 and 0818912.
References (55)
- et al.
EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis
J. Neurosci. Meth.
(2004) - et al.
Identifying the ERP correlate of a recognition memory search attempt
Brain Res. Cogn. Brain Res.
(2005) - et al.
Dissociable neural correlates for familiarity and recollection during the encoding and retrieval of pictures
Brain Res. Cogn. Brain Res.
(2004) - et al.
An electrophysiological study of scene effects on object identification
Brain Res. Cogn. Brain Res.
(2003) - et al.
Visual mental imagery and perception produce opposite adaptation effects on early brain potentials
Neuroimage
(2008) - et al.
Has the butcher on the bus dyed his hair? When color changes modulate ERP correlates of familiarity and recollection
Neuroimage
(2006) Neuroimaging studies of priming
Prog. Neurobiol.
(2003)- et al.
Priming, response learning and repetition suppression
Neuropsychologia
(2008) - et al.
The face-sensitive N170 and VPP components manifest the same brain processes: the effect of reference electrode site
Clin. Neurophysiol.
(2005) - et al.
Dissociating recollection from familiarity: electrophysiological evidence that familiarity for faces is associated with a posterior old/new effect
Neuroimage
(2007)
Scalp distributions of event-related potentials: an ambiguity associated with analysis of variance models
EEG Clin. Neurophysiol.
Validating neural correlates of familiarity
Trends Cogn. Sci.
Early lateralization and orientation tuning for face, word, and object processing in the visual cortex
Neuroimage
Event-related potentials and recognition memory
Trends Cogn. Sci.
Reductions in cortical activity during priming
Curr. Opin. Neurobiol.
Neurophysiological evidence for two processing times for visual object identification
Neuropsychologia
Object knowledge during entry-level categorization is activated and modified by implicit memory after 200 ms
Neuroimage
Activity in both hippocampus and perirhinal cortex predicts the memory strength of subsequently remembered information
Neuron
Conceptual and perceptual memory: retrieval orientations reflected in event-related potentials
Acta Psychol.
Context effects on the neural correlates of recognition memory: an electrophysiological study
Neuron
Neural substrates of remembering: electroencephalographic studies
Remembering and knowing: electrophysiological distinctions at encoding but not retrieval
Neuroimage
The nature of recollection and familiarity: a review of 30 years of research
J. Mem. Lang.
The neural basis of the butcher-on-the-bus phenomenon: when a face seems familiar but is not remembered
Neuroimage
Electrophysiological studies of human face perception: I. Potentials generated in occipitotemporal cortex by face and non-face stimuli
Cereb. Cortex
Grounded cognition
Annu. Rev. Psychol.
Recognition-by-components: a theory of human image understanding
Psychol. Rev.
Cited by (103)
Event-related potentials reveal visual episodic memory deficits in patients with temporal lobe epilepsy
2023, Epilepsy and BehaviorEvent-Related Potential (ERP) evidence for fluency and disfluency effects on recognition memory
2023, Brain and CognitionContext influences the FN400 recognition event-related potential
2020, International Journal of Psychophysiology