Elsevier

Neurobiology of Aging

Volume 34, Issue 3, March 2013, Pages 973-985
Neurobiology of Aging

Regular article
Event-related potentials dissociate perceptual from response-related age effects in visual search

https://doi.org/10.1016/j.neurobiolaging.2012.08.002Get rights and content

Abstract

Attentional decline plays a major role in cognitive changes with aging. However, which specific aspects of attention contribute to this decline is as yet little understood. To identify the contributions of various potential sources of age decrements in visual search, we combined response time measures with lateralized event-related potentials of younger and older adults performing a compound-search task, in which the target-defining dimension of a pop-out target (color/shape) and the response-critical target feature (vertical/horizontal stripes) varied independently across trials. Slower responses in older participants were associated with age differences in all analyzed event-related potentials from perception to response, indicating that behavioral slowing originates from multiple stages within the information-processing stream. Furthermore, analyses of carry-over effects from one trial to the next revealed repetition facilitation of the target-defining dimension and of the motor response—originating from preattentive perceptual and motor execution stages, respectively—to be independent of age. Critically, we demonstrated specific age deficits on intermediate processing stages when intertrial changes required more executively controlled processes, such as flexible stimulus-response (re-)mapping across trials.

Introduction

One essential daily task that becomes slower with age is visual search: our ability to discern and react upon a visually more or less distinctive item in a cluttered scene (Madden et al., 2004). Age-related slowing in the performance of visual search tasks might be attributable to a selective stage in the information-processing cycle, or it might originate from several stages and accrue across the sequential processes making up the cycle. Candidate stages are the selection of task-relevant sensory information, the identification of response-critical information, and/or the selection and execution of the required motor response (e.g., Kok, 2000, Salthouse, 2000). The present study was designed to examine this question at the microlevel of separable processing stages.

Performance in visual search tasks is known to be influenced by recently encountered stimuli and actions performed in response to them, within a sequence of trial episodes (e.g., Maljkovic and Nakayama, 1994, Müller et al., 2010). A special instance that might pose particular problems for older adults is the remapping of previously encoded stimulus-response (S-R) associations across such episodes (Hommel et al., 2011). Presumably, the fast adaptation processes involving flexible S-R remapping from one trial episode to the next require a higher degree of executive control processes, which are particularly age-sensitive (e.g., Park, 2000).

Using a cognitive neuroscience approach (e.g., Grady, 2008, Reuter-Lorenz and Park, 2010), we examined behavioral performance together with lateralized event-related potentials (ERP) of younger and older adults in order to determine the relative contributions of separable sources to age-related decrements in the performance of a visual pop-out search task. A pop-out target differs from distractors in a simple feature, providing a strong bottom-up signal for focal-attentional selection (Treisman and Gormican, 1988, Wolfe, 1994). Thus, this task is highly suitable for assessing the effects of aging, because the underlying cognitive architecture and the processing stages involved—perceptual selection, perceptual identification, response selection, response execution—are clearly defined (e.g., Müller and Krummenacher, 2006, Töllner et al., 2012b). Performance of a simple pop-out search task can be assumed to be relatively unaffected by strategies. This is a critical advantage compared to more complex search tasks, which are prone to search strategies that might systematically differ between younger and older adults (e.g., more careful search in older adults; see Hommel et al., 2004). Accordingly, age differences in pop-out search can be relatively unequivocally attributed to the well-defined processing stages involved.

Here, we used a variation of the classic pop-out paradigm, the so-called compound-search task (e.g., Bravo and Nakayama, 1992, Duncan, 1985), in which the selection- and the response-defining target features vary independently of each other across consecutive trials. The pop-out target was defined, variably across trials, by a unique feature in either the color dimension: a single red (striped) circle within a display of yellow (striped) circles, or in the shape dimension: a yellow (striped) square among yellow (striped) circles. Thus, target selection depended on detecting a salient color or, respectively, shape difference in the search array. Independently of this, the response was defined by the orientation of the pop-out target's stripes: “vertical” or “horizontal” orientation was responded to by a left or, respectively, right mouse button press (Fig. 1).

By analyzing carry-over effects from one trial to the next in this task (repetition vs. change of the target-defining dimension, repetition vs. change of the response-defining feature), we were able to compare the influence of recently encountered selection- and response-specific information between age groups. If a target on a given trial n is defined by the same feature or in the same dimension as the target on the preceding trial n − 1, attentional guidance is facilitated, due to featural/dimensional “priming” or “weighting” (e.g., Found and Müller, 1996, Maljkovic and Nakayama, 1994, Maljkovic and Nakayama, 2000; see also Töllner et al., 2009, for “modality-weighting”): responses are faster compared with when the search-critical feature or dimension changes across trials (e.g., a color target preceded by a color target vs. a color target preceded by a motion target). Note that featural/dimensional weighting operates largely implicitly and automatically. For instance, with regard to dimension weighting, this is evidenced by the fact the prioritization of the target-defining dimension on the preceding trial cannot be completely overcome by top-down control processes (Müller et al., 2003, Töllner et al., 2010) and is not dependent on explicit memory of the trial history (Müller et al., 2004). Critically, repetition facilitation has been shown to be largely spared from age-related decline (Kumada and Hibi, 2004, Madden et al., 2004, McCarley et al., 2004). However, in compound-search tasks, response times (RT) also vary with changes in the to-be-performed motor response across trials, more critically, with changes (vs. repetitions) of the S-R mapping (e.g., Lamy et al., 2010, Töllner et al., 2008). Although changes in the target-defining dimension and the response-defining feature are statistically independent of each other, an interactive RT pattern (e.g., Müller and Krummenacher, 2006) indicative of partial repetition costs (PRCs; e.g., Hommel, 2004) is typically observed: responses are faster when both the dimension- and response-defining attributes are repeated or when both change, and slower when only one of the two attributes changes while the other is repeated (e.g., Pollmann et al., 2006, Töllner et al., 2008). The magnitude of such PRCs has been suggested to depend on an individual's ability to flexibly break up the S-R association established on the previous trial and configure a new linkage—that is, essentially, the efficiency of executive control processes (Colzato et al., 2006, Hommel et al., 2011). It has previously been suggested that age-related changes in executive control processes, which are known to be particularly affected by aging, critically contribute to overproportional RT costs in elderly when stimulus-response remappings over trial sequences are required, as, for example, in the Simon Task (e.g., Castel et al., 2007) or in task-switching paradigms (e.g., Mayr, 2001). In particular, it has been proposed that control over response selection might be a crucial determinant of age effects (Hartley, 2001) and could be localized best by using tasks involving the control of more general or primitive sets of S-R mapping (Castel et al., 2007). On this basis, we tested whether PRCs in a simple compound-search task would also be particularly marked in older, in comparison with younger, adults.

By examining lateralized ERP responses in combination with RT measures (Posner, 2005), it becomes possible to disentangle the effects of aging, including their interactions with intertrial effects, on distinct stages of processing in visual search, in particular: those of (1) preattentive perceptual, (2) postselective perceptual, (3) response selection, and (4) response production processing (Mazza et al., 2009, Perron et al., 2009, Töllner et al., 2011b). Pursuing this approach (Fig. 2), we analyzed the following lateralized ERP components: the posterior contralateral negativity (PCN; this component is traditionally referred to as N2-posterior-contralateral; however, we prefer the term PCN to emphasize its independence from the nonlateralized N2; see, e.g., Shedden and Nordgaard, 2001, Töllner et al., 2011a); the sustained posterior contralateral negativity (SPCN; also referred to as contralateral delay activity; see Vogel and Machizawa, 2004); the stimulus-locked lateralized readiness potential (sLRP); and the response-locked lateralized readiness potential (rLRP).

The first parameter of interest, the PCN component, is a negative-going deflection elicited over lateral parieto-occipital sites contralateral to the location of an attended stimulus in the time window approximately 175–300 ms poststimulus (e.g., Luck and Hillyard, 1994, Woodman and Luck, 1999). The PCN is generally thought to reflect focal-attentional selection of task-relevant target objects among distracter items in visual space (e.g., Eimer, 1996, Woodman and Luck, 1999). It has been demonstrated that its latency varies markedly depending on a variety of top-down (e.g., featural task set: Eimer and Kiss, 2008; dimensional set: Töllner et al., 2010, Töllner et al., 2012a), bottom-up (e.g., stimulus intensity: Brisson et al., 2007; stimulus saliency: Töllner et al., 2011a), and intertrial factors (e.g., dimensional target identity of the previous trial: Töllner et al., 2008). Thus, given that the deployment of focal attention is guided by the outcome of early sensory feature-contrast computations, the timing of the PCN can be used as a temporal marker of the transition from the preattentive perceptual coding of the whole search array to the focal attentional processing of selected (target) stimulus (e.g., Luck et al., 2006). Recently, a delayed PCN has been found to index age-related slowing in visual attentional selection (Lorenzo-López et al., 2008).

Another lateralized posterior component manifesting at somewhat longer, 300–700 ms, poststimulus latencies (e.g., Jolicoeur et al., 2006, Perron et al., 2009), the SPCN, is assumed to reflect active visual short-term memory (vSTM) maintenance (e.g., Jolicoeur et al., 2006). This component is generated in visual attention tasks, including pop-out search (Mazza et al., 2007), which require detailed analysis of the selected target in vSTM (e.g., Dell'Acqua et al., 2006, Mazza et al., 2009). Thus far, the SPCN has not been examined for age-related changes in visual search tasks. However, in change detection tasks, especially in older age, the SPCN amplitude is modulated by the timing and efficiency of prior, attentionally controlled selection processes (Jost et al., 2011, Sander et al., 2011).

Finally, the lateralized readiness potential (LRP) is a negative-going deflection over the motor areas contralateral to the side of a unimanual response and has been linked to the activation and execution of effector-specific motor responses (e.g., Coles, 1989, Kutas and Donchin, 1980). The LRP onset calculated relative to stimulus onset (sLRP) indicates the point in time at which 1 of several possible responses is preferred—thus, reflecting the time required to initiate an effector-specific motor activation (i.e., response selection) after the completion of stimulus-response translation processes (e.g., Töllner et al., 2012b). By contrast, the LRP computed relative to response onset (rLRP) reflects the time demands required to produce and execute this response (e.g., Miller et al., 1998). Recently, age-related slowing in speeded RT tasks has been found to be related to prolonged and enhanced amplitudes of the rLRP—indicating that older adults require higher activation levels for motor execution, which are time-consuming to build up (Falkenstein et al., 2006, Yordanova et al., 2004).

In a pioneering study on young adults, Töllner et al. (2008) were able to attribute presumably automatic intertrial repetition facilitation and more controlled S-R (re-)mapping effects in a compound-search task to distinct substages of processing, based on combined PCN, sLRP, and rLRP analyses. The authors found, irrespective of motor response changes, repetitions (relative to changes) of the target-defining dimension to produce shorter latencies of the PCN, indicating that (at least part of) the dimension-specific intertrial facilitation originates at the preattentive stage of saliency coding. In turn, irrespective of dimension changes, repetitions (relative to changes) of the generated motor response reduced amplitudes of the rLRP, indicating that response-specific intertrial facilitation originates at the stage of motor-response production. Importantly, the sLRP was found to be modulated interactively by dimension and response changes versus repetitions: its onset latencies were delayed for partial repetitions, that is, when either only the dimension or only the response changed, compared with complete repetitions or changes. Thus, the sLRP latencies showed a PRC-type pattern, indicative of time-consuming S-R (re-)mapping processes at stages of response selection as the main source of the PRC effect found in the response times. Accordingly, assuming that S-R remapping across trials is particularly affected in older age, this should show up in alterations of this stage concerned with response selection (Fig. 2).

Taken together, the present study was designed to provide a complete and comprehensive picture of age-related changes in visual (pop-out) search by combining measures of (behavioral) response speed with those of event-related lateralizations. First, we examined separable stages of processing in task performance that might, selectively or additively, contribute to age-related slowing: (1) slower allocation of focal-attention in older compared with younger adults should be reflected in prolonged PCN latencies; (2) less effective target analysis in vSTM should be reflected in attenuated SPCN amplitudes; (3) retarded response selection should be reflected in delayed sLRPs (over and above any PCN latency difference); and (4) slowed motor response execution should be reflected in prolonged rLRPs. Second, we examined whether older adults would show particular decrements in S-R (re-)mapping across trials: based on previous reports with young adults (Töllner et al., 2008), we assumed increased behavioral PRCs in older adults to originate at processing stages concerned with stimulus-response transmission. In particular, finding more prolonged sLRP latencies for partial repetitions versus complete repetitions/changes in older, compared with younger, adults would argue that the age deficit is attributable to stages of response selection. Further, we examined whether SPCN amplitudes would be modulated by PRCs and age. A critical involvement of this component would support the functional interpretation relating the SPCN to the identification of response-critical target attributes during postselective maintenance of visual object information (e.g., Eimer and Kiss, 2010), indicative of postselective perceptual processing contributing to age-related decline in S-R (re-)mapping (Fig. 2).

Section snippets

Participants

Eighteen “young” and 18 “old” adults were included in the sample (Table 1). Two further older participants and one further younger participant were excluded from analyses due to excessive amounts of eye movement activity. Further exclusion criteria were any history of neurological (e.g., traumatic brain injury, stroke), psychiatric (e.g., depression, anxiety disorders), chronic somatic (e.g., hypertension, diabetes), and chronic eye diseases (e.g., glaucoma, cataract). All participants had

Raw response times

The ANOVA on RTs showed that, in general, older participants responded more slowly than younger participants (703 vs. 566 ms) (main effect Age: F(1,34) = 38.01; p < 0.001). Further, intertrial effects followed an interactive pattern, which varied with age (main effect Dimension Change: F(1,34) = 15.96; p < 0.001; Dimension Change × Response Change interaction: F(1,34) = 72.05; p < 0.001; Dimension Change × Response Change × Age interaction: F(1,34) = 8.81; p < 0.01). In particular, the PRC

Discussion

The present study was designed to examine age-related changes in visual search by a combined analysis of behavioral RTs and lateralized ERP components recorded while younger and older participants performed a compound search task. There was a marked general effect of age on behavioral response speed, in line with previous visual search studies (e.g., Hommel et al., 2004, Madden et al., 2004). Our results show that this RT cost is associated with age differences at all dissociable substages of

Disclosure statement

There are no actual or potential conflicts of interest.

All participants gave informed consent and received payment for participating. Approval was obtained from the Ethics Committee of the Faculty of Psychology, LMU Munich. Written informed consent according to the Declaration of Helsinki II was obtained from all participants. All participants received payment.

Acknowledgements

This research was supported by grants from DFG Excellence Cluster EC 142 “CoTeSys” (HJM) and DFG research group FOR480 (HJM). I.W. receives a scholarship from the Bavarian Elite Fund (BayEFG).

References (83)

  • C. Lustig et al.

    Preserved neural correlates of priming in old age and dementia

    Neuron

    (2004)
  • H.J. Müller et al.

    Dimension-based attention modulates feed-forward visual processing

    Acta Psychol. (Amst.)

    (2010)
  • S. Pollmann et al.

    Selective and interactive neural correlates of visual dimension changes and response changes

    NeuroImage

    (2006)
  • T.A. Salthouse

    Aging and measures of processing speed

    Biol. Psychol

    (2000)
  • J.M. Shedden et al.

    ERP time course of perceptual and post-perceptual mechanisms of spatial selection

    Brain Res. Cogn. Brain Res

    (2001)
  • T. Töllner et al.

    Top-down weighting of visual dimensions: Behavioral and electrophysiological evidence

    Vision Res

    (2010)
  • N. Wild-Wall et al.

    Flanker interference in young and older participants as reflected in event-related potentials

    Brain Res

    (2008)
  • Guideline thirteen: guidelines for standard electrode position nomenclatureAmerican Electroencephalographic Society

    J. Clin. Neurophysiol

    (1994)
  • M.J. Bravo et al.

    The role of attention in different visual-search tasks

    Percept. Psychophys

    (1992)
  • J.F. Brinley

    Cognitive sets, speed and accuracy of performance in the elderly

  • B. Brisson et al.

    Stimulus intensity affects the latency but not the amplitude of the N2pc

    Neuroreport

    (2007)
  • A.D. Castel et al.

    Spatial attention and response control in healthy younger and older adults and individuals with Alzheimer's disease: evidence for disproportionate selection breakdowns in the Simon task

    Neuropsychol

    (2007)
  • J. Cerella

    Generalized slowing in Brinley plots

    J. Gerontol

    (1994)
  • M.G. Coles

    Modern mind-brain reading: psychophysiology, physiology, and cognition

    Psychophysiology

    (1989)
  • L.S. Colzato et al.

    Intelligence and cognitive flexibility: fluid intelligence correlates with feature “unbinding” across perception and action

    Psychon. Bull. Rev

    (2006)
  • N. Cowan et al.

    Life-span development of visual working memory: when is feature binding difficult?

    Dev. Psychol

    (2006)
  • I.J. Deary et al.

    The neuroscience of human intelligence differences

    Nat. Rev. Neurosci

    (2010)
  • R. Dell'Acqua et al.

    Spatial attention freezes during the attention blink

    Psychophysiology

    (2006)
  • J. Duncan

    Visual search and visual attention

  • M. Eimer et al.

    Involuntary attentional capture is determined by task set: evidence from event-related brain potentials

    J. Cogn. Neurosci

    (2008)
  • M. Eimer et al.

    An electrophysiological measure of access to representations in visual working memory

    Psychophysiology

    (2010)
  • J.H. Fecteau et al.

    Exploring the consequences of the previous trial

    Nat. Rev. Neurosci

    (2003)
  • A. Found et al.

    Searching for unknown feature targets on more than one dimension: investigating a “dimension-weighting” account

    Percept. Psychophys

    (1996)
  • A. Gazzaley et al.

    Age-related top-down suppression deficit in the early stages of cortical visual memory processing

    Proc. Natl. Acad. Sci. U. S. A.

    (2008)
  • C.L. Grady

    Cognitive neuroscience of aging

    Ann. N. Y. Acad. Sci

    (2008)
  • K. Gramann et al.

    Brain electrical correlates of dimensional weighting: an ERP study

    Psychophysiology

    (2007)
  • K. Gramann et al.

    Dimension-based attention modulates early visual processing

    Psychophysiology

    (2010)
  • A.A. Hartley

    Age differences in dual-task interference are localized to response-generation processes

    Psychol. Aging

    (2001)
  • A.P. Hillstrom

    Repetition effects in visual search

    Percept. Psychophys

    (2000)
  • B. Hommel et al.

    Feature integration across the lifespan: stickier stimulus-response bindings in children and older adults

    Front. Psychol

    (2011)
  • B. Hommel et al.

    Visual search across the life span

    Dev. Psychol

    (2004)
  • Cited by (0)

    View full text