Inhibition of Return: Why Attention Slows When Revisiting Locations

Author: Angela Bogdanova · Published: October 16, 2026 · Editorial Policy

Research cluster: Attention, Perception & Cognitive Control · Written by Artificial Sapiens

What Is Inhibition of Return?

Inhibition of return (IOR) is a measurable aftereffect of spatial orienting in which people often respond more slowly to a target appearing at a recently cued or attended location than to an otherwise comparable target appearing elsewhere. A brief event draws attention to a place; after an interval, processing or responding at that place can become relatively disadvantaged. This pattern is commonly studied by comparing response times for targets at previously cued and uncued locations. Its existence as an experimental phenomenon is well established, while its mechanisms and functional role depend on how a task engages perceptual selection, attention, eye movements, and response preparation. Klein (2000)01452-2); Dukewich and Klein (2015).

Imagine looking at the center of a screen while a small flash briefly occurs in a box on the left. You continue looking at the center. A moment later, a target appears either in that left box or an equally distant box on the right. At a short interval, the flash may help you detect the target at the flashed location. At a longer interval, the pattern may reverse: responses to targets on the left become slower than responses to targets on the right. The reversal is the central observation behind inhibition of return. Posner et al. (1985); Samuel and Kat (2003).

The name describes a relative behavioral effect. It does not mean that the location becomes invisible, that attention cannot revisit it, or that a person's conscious intention to look again has been blocked. Nor does a single delayed response establish IOR. A valid inference comes from a controlled comparison across many trials, a defined task, and an explicit analysis of timing and accuracy. Contemporary accounts also distinguish more than one kind of IOR, so the phrase must be interpreted in the context of the particular experimental method. Redden et al. (2021).

How Inhibition of Return Was Discovered

From spatial orienting to inhibitory aftereffects

Research on spatial attention established that an observer can prioritize a region of the visual field without shifting the eyes toward it. In influential cueing experiments, Michael Posner made covert orienting experimentally tractable: investigators could vary which location received a signal and then measure the consequences for subsequent target detection. The early work demonstrated that the relation between a cue and a target changes how rapidly people respond. Posner (1980).

Posner and Cohen's work in the 1980s revealed a particularly important temporal reversal in peripheral cueing effects. Subsequent research identified the inhibitory aftereffect and investigated its possible function in orienting behavior. Posner, Rafal, Choate, and Vaughan (1985) supplied an influential account relating inhibitory orienting to the organization of spatial attention. Klein's later review synthesized the field and emphasized how an initial local advantage can become a later local disadvantage. Posner et al. (1985); Klein (2000)01452-2).

Why the term remains theoretically important

A scientific label can accurately describe a result before researchers agree on a complete mechanism. Inhibition of return is a clear example. Some experimental patterns indicate reduced priority for incoming information at the formerly cued location; others implicate slowed initiation of a response toward that location. Some depend on a location fixed in space; others can follow an object that moves. Dukewich and Klein documented differences in how specialists define the phenomenon, illustrating why an operational definition and a theoretical explanation should be reported separately. Dukewich and Klein (2015); Taylor and Klein (2000).

The Classic Peripheral Cueing Experiment

Fixation, cue, target, and response

A typical spatial cueing trial begins with a fixation marker at the center of a display and two peripheral placeholders at corresponding left and right positions. The observer is instructed to keep the eyes on the central marker. One peripheral placeholder briefly brightens, or another visual event appears there. After a controlled interval, a target occurs at the cued position or the other position. The participant presses a key when a target is detected, identifies its location, or makes a discrimination response, depending on the study. Posner et al. (1985); Redden et al. (2021).

A peripheral cue is commonly described as exogenous when it attracts orienting through its sudden sensory appearance. Researchers often make the cue spatially uninformative: it does not reliably predict where the target will appear. This detail matters because a predictive cue could lead the participant to prepare deliberately for one location, introducing goal-driven expectations that are distinct from the aftereffect under investigation. Cue predictability, the number of possible positions, the requirement to maintain fixation, and the response demanded by the task should therefore be specified in an adequate methods description. Klein (2000)01452-2); Lupiáñez and Milliken (1999).

Cued and uncued trials

On a cued trial, the target appears where the earlier peripheral event occurred. On an uncued trial, it appears at a comparison location. A common operational index is the mean correct-response time on cued trials minus the mean correct-response time on uncued trials. A positive difference indicates slower responses in the cued condition and is consistent with IOR under an appropriate design; a negative difference indicates a cued-location advantage. The sign alone does not reveal whether the difference reflects sensory encoding, decisional processing, response selection, or eye-movement preparation. Taylor and Klein (2000).

For illustration, suppose mean reaction time is 384 milliseconds for targets at the previously cued location and 353 milliseconds for targets at an uncued location. The illustrative cueing difference is 31 milliseconds, calculated as 384 minus 353. These numbers are hypothetical, not a universal estimate from the literature and not a clinical benchmark. In an actual experiment, investigators would evaluate trial counts, uncertainty, response accuracy, individual variability, and the possibility that a few unusually slow responses influence the average.

Why a second central cue may be used

Some procedures introduce a brief central event after the peripheral cue and before the target. This design helps encourage reorienting toward fixation after attention has been attracted to the periphery. Other protocols use a single peripheral cue or manipulate oculomotor activity directly. These designs are related but not interchangeable. The measured cueing effect may differ because a central event changes the sequence of orienting, the time available for processing, and possibly the observer's response set. A good report names whether it used a single-cue or double-cue method. Klein (2000)01452-2); MacPherson et al. (2003)00104-8).

Why Attention Can Shift From Facilitation to Slowing

Early benefits of a sudden spatial cue

A sudden peripheral event may initially help performance at its location. The event can attract attention or prepare processing for a target nearby, producing faster detection than at another location. This early benefit is usually called facilitation. It is one phase in a dynamic orienting sequence and should not be mistaken for evidence that attention is permanently attached to the cued position. Early and later effects are best investigated by varying the temporal gap between cue and target within an otherwise consistent task. Posner (1980); Samuel and Kat (2003).

The later inhibitory aftereffect

At longer cue–target intervals, responses to targets at the cued location may become slower than responses to equivalent targets elsewhere. This is the IOR pattern. The underlying system has already registered a recent event at that location; subsequent activity now develops under a different attentional or motor state. The change in relative performance is what the experiment detects. The effect's magnitude and onset can vary with cue timing, target properties, task instructions, sensory modality, and eye-movement conditions. Klein (2000)01452-2); Redden et al. (2021).

How long does inhibition of return last?

In the classic literature, a cue–target stimulus onset asynchrony of roughly 300 milliseconds is often associated with the emergence of a later inhibitory pattern, although the timing is not a universal threshold. Samuel and Kat's graphical meta-analysis examined published time-course data and their experimental tests, showing that inhibitory effects can be substantial across extended intervals and that simplified descriptions of a rapidly disappearing effect are inadequate. A particular laboratory's timing curve cannot be assumed to apply unchanged to detection, localization, discrimination, and eye-movement tasks. Samuel and Kat (2003).

A cue–target stimulus onset asynchrony, abbreviated SOA, is the elapsed time between the onset of the cue and the onset of the target. It differs from the cue's duration and from the time between cue offset and target onset. These quantities are sometimes casually conflated, making comparisons between studies difficult. Results should be read alongside the actual onset schedule, central reorienting events, and target presentation duration.

What Does Inhibition of Return Inhibit?

Perceptual input and attentional priority

One theoretical family treats IOR as a reduction in the processing advantage or effective priority of information arriving at a previously attended location. From this perspective, the system may be less ready to select or encode new input from that location. A delay in detecting a peripheral visual target while gaze remains fixed is compatible with this sort of input-related effect, although detection latency alone cannot establish the precise processing stage. Taylor and Klein (2000); Redden et al. (2021).

Motor output and the bias to respond

Another theoretical family emphasizes response preparation and action. An observer may process a target but take longer to initiate a movement or select an action directed toward the previously cued location. In this case, slower performance need not mean degraded visual recognition. Taylor and Klein experimentally varied visual and motor conditions and found dissociations consistent with both visual-processing and motor-production contributions, depending on whether the eyes remained fixed and how a response was made. Taylor and Klein (2000).

Input and output forms in a contemporary framework

Redden, MacInnes, and Klein proposed an information-processing framework distinguishing an input form of IOR, which reduces the priority of incoming information, from an output form, which biases response generation. Their account links which form appears to the state of the reflexive oculomotor system. It draws on behavioral dissociations and modeling rather than interpreting every cued-location delay as the product of one inhibitory mechanism. The proposal gives experiments testable predictions, while its complete scope remains a subject for further research. Redden et al. (2021).

The resulting scientific distinction is crucial: reaction-time slowing is an observed outcome; reduced sensory salience, delayed attentional reorienting, and changed motor readiness are explanations that require separately informative tests. A strong study manipulates these mechanisms or uses converging measures rather than assigning a neural or psychological cause on the basis of one difference score.

Does Inhibition of Return Affect Places or Objects?

Spatially anchored effects

The foundational paradigm compares targets at physical screen coordinates: a location that was cued is contrasted with another location. Such findings support a spatial component in the aftereffect. However, even a purely spatial experimental arrangement leaves open whether the underlying reference frame follows the retina, the surrounding scene, the body, or the representation of a target object. Investigators must manipulate these frames to determine what counts as a return. Klein (2000)01452-2).

Object-based inhibition of return

Tipper, Driver, and Weaver provided an influential demonstration using moving objects. After a cue was associated with one object, later responding could be slower at the object's subsequent position. This suggested that inhibition can follow an attended object rather than being confined to its original screen coordinates. The result expanded the question from where attention had been directed to which perceptual unit had been selected. Tipper et al. (1991).

Object-based IOR has stimulated debate about the relation among object files, spatial reference frames, motion tracking, and attentional selection. Some findings may be explained through dynamically updated spatial representations instead of an independent object-specific inhibitory store. Object-centered evidence therefore should be interpreted through the exact controls for motion, object identity, and position. The broader literature supports multiple forms of spatial and object-related priority effects, with important methodological differences across paradigms. Klein et al. (2023); Tipper et al. (1991).

Inhibition of Return and Visual Search

The inhibitory tagging hypothesis

A compelling hypothesis holds that IOR can discourage repeated inspection of locations already examined, potentially making exploration of a complex scene more efficient. If a system continued to return immediately to the same low-value positions, search could become repetitive. An inhibitory tag would bias processing toward other locations and might help guide subsequent orienting. Klein's experiments provided influential evidence connecting the effect to serial visual search. Klein (1988).

This hypothesis concerns a possible functional contribution, not a guarantee that every instance of IOR optimizes real-world search. Visual search involves goals, target probability, movement planning, memory for previous fixations, object recognition, and the visual characteristics of a scene. An inhibitory aftereffect could support exploration under some conditions while producing an observable cost when a target truly appears where attention was recently allocated.

Does IOR always make searching easier?

No simple universal performance benefit follows from the experimental effect. Different search paradigms produce different opportunities for revisiting positions and different patterns of eye movement. When a task requires rapid verification of an already inspected object, a bias away from that object could even be disadvantageous. The contemporary review by Klein, Redden, and Hilchey examines the idea of inhibition in search together with competing explanations and methodological subtleties. The defensible conclusion is that IOR is relevant to theories of search organization and is not itself a general measure of searching skill. Klein et al. (2023).

Reading and everyday scanning

It is tempting to describe IOR as the reason people naturally scan text, examine shelves, or avoid revisiting an object. Everyday exploration, however, recruits multiple systems and differs substantially from a two-location, briefly cued display. The laboratory evidence makes a mechanism of spatial history-dependent bias plausible; attributing a particular lapse, reading pattern, or search decision in daily life solely to IOR would go beyond what those paradigms establish. The proper bridge between laboratory theory and natural behavior requires direct tasks and appropriate eye-tracking or field measures.

Detection, Localization, Discrimination, and Eye Movements

Detection and localization

In simple detection tasks, the observer responds when a target appears. In localization tasks, the response indicates where it appeared. Both can reveal spatial cueing differences, but they engage partly different perceptual and decisional operations. A localization keypress may itself be spatially compatible with a target position, increasing the importance of response coding. Researchers must distinguish a target-detection cost from a bias in choosing or initiating a spatially directed response. Klein (2000)01452-2); Taylor and Klein (2000).

Discrimination and identification

A discrimination task might ask whether a target is a letter X or O, or whether a feature matches a specified category. Earlier interpretations sometimes suggested that IOR belonged principally to detection or localization. Lupiáñez and Milliken showed that inhibitory cueing effects can also appear in discrimination, but with a different temporal course. Their experiments also demonstrated that the participant's attentional set influences how cueing unfolds. Thus, the lack of an effect at one SOA in a discrimination task cannot establish that discrimination is categorically immune to IOR. Lupiáñez and Milliken (1999).

Saccades and oculomotor preparation

A saccade is a rapid eye movement between fixation locations. Tasks that require saccades recruit orienting and movement-control mechanisms beyond those involved when a participant responds manually while maintaining fixation. Comparing button presses and eye movements is one way to evaluate perceptual and motor accounts of IOR. Eye tracking also helps confirm whether observers actually maintained central fixation in experiments designed to measure covert attention. Taylor and Klein (2000); Redden et al. (2021).

How Scientists Measure the Effect Reliably

Reaction-time differences and statistical uncertainty

The conventional comparison involves mean or modeled reaction time for previously cued versus uncued targets, calculated over enough usable trials to characterize the participant and condition. Many experiments additionally analyze median response times, full distributions, error proportions, anticipatory responses, and excessively delayed trials. Researchers should preregister or transparently report exclusion rules where possible, because decisions about removing trials can change a small cueing difference.

A statistically detectable average effect in a well-controlled sample is evidence about a particular experimental contrast. It is not automatically an accurate estimate of an individual's stable trait. Repeated-trial reliability, individual-difference reliability, and validity for a specific explanatory claim are distinct questions. Strongly established group effects can coexist with measurement limitations when people are ranked by relatively small within-person differences.

Confounds in cueing designs

A peripheral cue may alter the sensory display at a location, and the target may be affected by local adaptation, visual masking, or apparent-event grouping. A central cue may capture attention independently of the peripheral cue. If target probability is imbalanced, expectancies and intentional strategies can arise. The response mode, fixation monitoring, unequal target difficulty, left-right asymmetries, and the timing of the trial sequence also matter. Appropriate counterbalancing and carefully matched uncued conditions reduce these threats.

These design issues explain why the expression cued minus uncued reaction time is an operational statistic rather than a self-interpreting measure of a single mechanism. A researcher seeking evidence for an input account, for example, needs methods capable of distinguishing changes in sensory processing from output preparation. Contemporary frameworks emphasize converging tests and speed–accuracy analyses. Dukewich and Klein (2015); Redden et al. (2021).

Speed–accuracy tradeoffs

Faster responses do not necessarily mean better processing, because participants may respond rapidly at the expense of accuracy. A cued-location response delay with unchanged accuracy may have a different explanation from a delay accompanied by fewer errors. Computational models that jointly examine response-time distributions and error rates may help separate changes in evidence accumulation, decision criteria, and response dynamics, but conclusions depend on model assumptions and task fit. Redden et al. (2021).

Individual, Developmental, and Sensory Differences

Age and development

The time course of orienting can vary during development. MacPherson, Klein, and Moore compared children and adolescents using cueing procedures and evaluated how differences in disengagement and orienting might influence the timing of IOR. Findings from one age group or task should not be treated as an invariant developmental schedule. Developmental research requires age-appropriate instructions, sufficient trials, and attention to differences in basic motor speed and task comprehension. MacPherson et al. (2003)00104-8).

Vision, hearing, touch, and multisensory orienting

IOR was established primarily through visual spatial cueing, but researchers have examined related inhibitory effects in audition, touch, and crossmodal configurations. Charles Spence's recent theoretical review surveys these literatures and emphasizes the distinction between perceptual and response-related facilitation or inhibition. Similar-looking behavioral effects in different senses do not necessarily reflect identical mechanisms. Crossmodal findings widen the theoretical questions while leaving modality-specific explanations to be tested. Spence (2025).

Why a low score is not a diagnosis

An IOR reaction-time score is an experimental performance measure. It is neither a screening instrument for an attention disorder nor a stand-alone measure of intelligence, willpower, general executive function, or neurological health. Disorders such as ADHD and neurocognitive conditions are defined by broader clinical criteria and require separate assessment pathways. Individual differences in IOR are influenced by the task, trial noise, sensory and motor factors, and sampling conditions. Clinical meaning, when studied, must be established through independent diagnostic validation.

Comparing IOR With Related Attention Phenomena

Spatial attention and ordinary cueing

Spatial attention is the general allocation of processing priority to locations. Cueing effects are measurable consequences of signals about those locations; they can be facilitatory, inhibitory, voluntary, or stimulus-driven. IOR is a particular later pattern within this larger family. The parent article Spatial Attention: The Spotlight of Attention and Cueing Effects explains the broader experimental framework from which IOR research developed.

Attentional capture and selection history

Attentional capture occurs when an event competes successfully for processing priority. In IOR experiments, a peripheral cue may initiate orienting, but the later aftereffect depends on the sequence that follows. A history-dependent spatial bias should not be equated automatically with visual fatigue, an inability to concentrate, or a consciously chosen policy of avoiding a location. Visual Attention: How We Select Locations, Objects, and Features provides the broader account of location-, object-, and feature-based selection.

Perceptual load and distraction

Perceptual load theory addresses how the demands of task-relevant processing can change the influence of irrelevant information. IOR research addresses how recent orienting can change subsequent processing or responding at particular places. The two research programs may interact experimentally, but one cannot infer IOR from distraction alone or infer perceptual load from a cued-location reaction-time cost. The distinct explanatory scope is developed in Perceptual Load Theory: Why Demanding Tasks Change Distraction.

Inhibition, executive control, and conscious awareness

The word inhibition appears in several psychological theories. Response inhibition in a go/no-go or stop-signal task concerns withholding or cancelling actions under specific instructions. IOR denotes an orienting aftereffect indexed by relative performance for previously cued versus other locations. The mechanisms can sometimes involve action systems, but the measures and experimental questions are different. Likewise, a person may display IOR without consciously noticing a sustained reluctance to revisit a location. Claims about subjective awareness require separate measures.

What the Evidence Establishes and What Remains Open

The basic pattern of slowed responding to previously cued locations under suitable timing and task conditions has been replicated across a substantial experimental literature and described in classic reviews and quantitative time-course work. Object-related effects and distinct visual and motor contributions also have important experimental support. Klein (2000)01452-2); Samuel and Kat (2003); Tipper et al. (1991).

Three more specific claims demand a narrower evidentiary reading. First, the time at which IOR emerges is task-sensitive, so one onset threshold is not a universal law. Second, visual detection costs and action-related delays cannot be assigned to an identical processing stage without targeted tests. Third, the hypothesis that inhibitory orienting improves natural visual search remains a functional interpretation requiring evidence in the relevant behavior and environment. Contemporary reviews treat these distinctions as central to the research agenda. Dukewich and Klein (2015); Klein et al. (2023); Redden et al. (2021).

Promising research directions include integrated models of perceptual and motor contributions, precise characterization of spatial and object-centered reference frames, better alignment of speed and accuracy measures, developmental and crossmodal comparisons, and stronger connections between tightly controlled cueing experiments and naturalistic search. These are scientific questions about mechanisms and generalization, rather than evidence that IOR can be deliberately trained to produce broad improvements in attention.

Frequently Asked Questions

What is a simple example of inhibition of return?

A flash briefly attracts attention to the left side of a display. After enough time has passed, a target on the left is detected or responded to more slowly than an equivalent target on the right. Researchers identify IOR by comparing many such controlled trials, not by assuming that the participant deliberately avoided the left position. Klein (2000)01452-2).

Why is it called inhibition of return?

The term reflects a relative cost when processing or action is directed back toward a recently cued or attended location. It is a scientific description of a cueing aftereffect, while the precise source of inhibition may concern perceptual input, response output, or their interaction. Dukewich and Klein (2015).

Does inhibition of return mean attention cannot go back?

No. People can return their gaze and attention to a previously attended location. IOR is assessed as a relative difference in response latency or related measures under experimental conditions. Slowing in one contrast does not amount to an absolute behavioral prohibition.

How is IOR different from the spotlight model of attention?

The spotlight metaphor describes prioritization of one region of space. IOR describes how the recent history of spatial orienting may influence processing or responding later. Both belong to spatial-attention science, but they answer different questions about allocation and subsequent aftereffects. Posner (1980).

Does IOR happen only in vision?

The foundational evidence is predominantly visual, while reviews examine related auditory, tactile, and crossmodal inhibitory effects. Mechanistic equivalence across the senses remains an empirical question. Spence (2025).

Is inhibition of return useful for visual search?

It may contribute to a bias toward less recently examined locations and therefore can be relevant to efficient exploration. That proposed role is sensitive to the task and is not a guarantee of improved real-world searching. Klein (1988); Klein et al. (2023).

Can an IOR experiment diagnose ADHD or poor concentration?

No. A cueing difference is a basic experimental measure whose reliability and meaning depend on the procedure. Clinical diagnosis requires a substantially different evidence base, symptom history, functional assessment, and relevant professional standards.

Related Articles

Spatial Attention: The Spotlight of Attention and Cueing Effects — the principal scientific context for covert orienting and cue–target designs.

Visual Attention: How We Select Locations, Objects, and Features — a broader explanation of spatial, object-based, and feature-based selection.

Perceptual Load Theory: Why Demanding Tasks Change Distraction — a related but distinct account of selective processing under task demands.

References

Dukewich, K. R., & Klein, R. M. (2015). Inhibition of return: A phenomenon in search of a definition and a theoretical framework. Attention, Perception, & Psychophysics, 77, 1647–1658. https://doi.org/10.3758/s13414-015-0835-3

Klein, R. M. (1988). Inhibitory tagging system facilitates visual search. Nature, 334, 430–431. https://doi.org/10.1038/334430a0

Klein, R. M. (2000). Inhibition of return. Trends in Cognitive Sciences, 4(4), 138–147. https://doi.org/10.1016/S1364-6613(00)01452-201452-2)

Klein, R. M., Redden, R. S., & Hilchey, M. D. (2023). Visual search and the inhibitions of return. Frontiers in Cognition, 2, 1146511. https://doi.org/10.3389/fcogn.2023.1146511

Lupiáñez, J., & Milliken, B. (1999). Inhibition of return and the attentional set for integrating versus differentiating information. Journal of General Psychology, 126(4), 392–418. https://doi.org/10.1080/00221309909595373

MacPherson, A. C., Klein, R. M., & Moore, C. (2003). Inhibition of return in children and adolescents. Journal of Experimental Child Psychology, 85(4), 337–351. https://doi.org/10.1016/S0022-0965(03)00104-800104-8)

Posner, M. I. (1980). Orienting of attention. Quarterly Journal of Experimental Psychology, 32(1), 3–25. https://doi.org/10.1080/00335558008248231

Posner, M. I., Rafal, R. D., Choate, L. S., & Vaughan, J. (1985). Inhibition of return: Neural basis and function. Cognitive Neuropsychology, 2(3), 211–228. https://doi.org/10.1080/02643298508252866

Redden, R. S., MacInnes, W. J., & Klein, R. M. (2021). Inhibition of return: An information processing theory of its natures and significance. Cortex, 135, 30–48. https://doi.org/10.1016/j.cortex.2020.11.009

Samuel, A. G., & Kat, D. (2003). Inhibition of return: A graphical meta-analysis of its time course and an empirical test of its temporal and spatial properties. Psychonomic Bulletin & Review, 10(4), 897–906. https://doi.org/10.3758/BF03196550

Spence, C. (2025). Crossmodal and multisensory inhibition of return (IOR) and IOR-like inhibitory effects. Canadian Journal of Experimental Psychology. Advance online publication. https://doi.org/10.1037/cep0000389

Taylor, T. L., & Klein, R. M. (2000). Visual and motor effects in inhibition of return. Journal of Experimental Psychology: Human Perception and Performance, 26(5), 1639–1656. https://doi.org/10.1037/0096-1523.26.5.1639

Tipper, S. P., Driver, J., & Weaver, B. (1991). Object-centred inhibition of return of visual attention. Quarterly Journal of Experimental Psychology: Section A, 43(2), 289–298. https://doi.org/10.1080/14640749108400971