Gollin Figures Test

Abstract

The Gollin Figures Test (often referred to as the Gollin Incomplete Figures Test or GFT) is a classic neuropsychological and psychophysical assessment instrument designed to measure visual perceptual closure, visual perceptual organization, and non-declarative implicit memory (specifically visual perceptual repetition priming). Developed by developmental psychologist Eugene S. Gollin in 1960, the test presents respondents with fragmented line drawings of everyday objects across graduated levels of perceptual completion. The classic standardized battery consists of 11 sets of line drawings, each organized into five ascending hierarchical levels of visual degradation—ranging from heavily fragmented, disjointed contours (Level 1) to fully intact, continuous line illustrations (Level 5). Administration typically follows an ascending method of limits: respondents view cards in order of increasing visual information until correct object identification is achieved, establishing an initial perceptual recognition threshold. Following an experimental delay ranging from minutes to days or weeks, the test is readministered to evaluate repetition priming, quantified as the reduction in visual information required for identification upon re-exposure (perceptual savings).

The Gollin Figures Test holds a central place in the history of cognitive neuroscience. Historically landmark investigations in amnesic patients—most notably Patient H.M. (Milner, Corkin, & Teuber, 1968) and Korsakoff syndrome cohorts (Warrington & Weiskrantz, 1968)—demonstrated intact learning and perceptual savings across repeated administrations of the Gollin figures despite profound, debilitating anterograde amnesia and zero conscious recollection of previous exposure. Psychometrically, the instrument evaluates both visual-perceptual closure thresholds (associated with ventral visual stream integrity in the occipitotemporal cortex) and pre-semantic perceptual priming (mediated by extrastriate cortical adaptation independent of medial temporal lobe structures). While initial versions relied on physical paper card series, contemporary psychometrics utilizes digital tachistoscopic and computerized variants with high temporal resolution and continuous degradation vectors. Psychometric analyses reveal robust split-half reliability, strong discriminant validity between cortical visual agnosias and medial temporal amnesias, and distinctive impairment patterns in neurodegenerative diseases such as Alzheimer’s disease, dementia with Lewy bodies, and Parkinsonian neurocognitive disorders.

Keywords

Gollin Figures Test, incomplete figures, implicit memory, perceptual priming, visual closure, repetition priming, visual agnosia, neuropsychological assessment, non-declarative memory, Gestalt perception, perceptual learning, visual object recognition

Authors

The Gollin Figures Test was originated, designed, and empirically operationalized by Eugene S. Gollin (Ph.D.), an influential American developmental and comparative psychologist whose career spanned foundational research in developmental perception, learning theory, and evolutionary behavioral analysis. Gollin completed his doctoral training in psychology and conducted extensive experimental research across academic institutions, notably serving as Professor of Psychology at Queens College of the City University of New York (CUNY) and later as Professor Emeritus of Psychology at the University of Colorado Boulder.

Gollin introduced the incomplete figures paradigm in his seminal 1960 monograph, “Preparing visual materials for developmental study: The incomplete figures test”, published in Perceptual and Motor Skills, followed by systematic developmental investigations examining developmental differences in perceptual recognition thresholds across preschool children, school-age youth, and adults (Gollin, 1961, 1962, 1965). Gollin’s central objective was not initially the diagnosis of memory pathology, but rather the psychophysical and developmental charting of how human observers synthesize fragmented sensory inputs into coherent conceptual representations. Subsequent clinical adaptations, standardizations, and neurological translations were pioneered by neuropsychologists including Brenda Milner, Suzanne Corkin, Elizabeth K. Warrington, and Lawrence Weiskrantz, who adapted Gollin’s stimulus cards into formal clinical neuropsychological batteries.

Purpose

The primary clinical and experimental purpose of the Gollin Figures Test is the quantitative assessment of two distinct, dissociable neuropsychological domains: (1) visual perceptual organization and closure, defined as the visual system’s capacity to integrate degraded, fragmented contours into an identifiable holistic representation (structural description system); and (2) perceptual repetition priming, a major subtype of implicit (non-declarative) memory wherein previous exposure to a perceptual stimulus facilitates its subsequent identification under conditions of perceptual degradation, without requiring conscious, intentional episodic recollection.

In clinical neuropsychology, the Gollin Figures Test serves critical diagnostic and localized lesion-mapping functions:

• Dissociation of Memory Subsystems: The test provides a gold-standard behavioral paradigm for distinguishing episodic memory impairment (governed by the hippocampus and medial temporal lobe structures) from perceptual representation systems (PRS) mediated by the occipitotemporal neocortex. Patients with profound episodic memory loss secondary to bilateral hippocampal sclerosis, anoxia, or early-stage Alzheimer’s disease demonstrate preserved perceptual savings on the Gollin Figures Test, identifying items at significantly lower degradation levels on Day 2 compared to Day 1, despite having complete episodic amnesia for the testing room, administrator, and testing event.
• Detection of Apperceptive and Associative Visual Agnosia: For patients suffering from cerebrovascular accidents, traumatic brain injuries, or posterior cortical atrophy (PCA), initial exposure baseline thresholds reveal severe impairments in visual feature integration. Patients with apperceptive visual agnosia fail to recognize Gollin figures until they reach complete or nearly complete presentations (Levels 4 or 5), or fail altogether, reflecting an inability to bind discontinuous visual contours into canonical structural descriptions.
• Differential Diagnosis in Neurodegenerative Conditions: Visual priming performance on fragmented figure tests provides valuable differential markers between neurodegenerative profiles. For example, patients with Dementia with Lewy Bodies (DLB) frequently display disproportionate deficits in baseline perceptual closure and elevated identification thresholds relative to amnesic Alzheimer’s patients, reflecting posterior parietal-occipital hypometabolism and visual processing pathobiology.
• Pediatric Neurodevelopmental Evaluation: In developmental psychology and pediatric neuropsychology, the test tracks the developmental maturation of visual closure, spatial integration, and perceptual speed in children with neurodevelopmental disorders, specific learning disabilities, autism spectrum conditions, and fetal alcohol spectrum disorders.

Psychological Construct

The psychological architecture measured by the Gollin Figures Test comprises a multi-layered interaction between low-level visual parsing, intermediate Gestalt feature binding, high-level object representation access, and non-conscious mnemonic retention:

1. Visual Perceptual Closure

Visual perceptual closure represents the visual cognitive system’s ability to deduce a meaningful whole from incomplete or fragmented sensory parts. When an observer views Level 1 of a Gollin figure (e.g., an umbrella rendered with 80% to 90% of its lines deleted), retinal input provides only discrete, disconnected line segments and spatial discontinuities. The visual cognitive apparatus must execute rapid spatial grouping, collinearity analysis, illusory contour completion, and figure-ground segmentation. This process relies heavily on the ventral visual processing pathway, progressing from primary visual cortex (V1/V2) through lateral occipital complexes (LOC) to the inferior temporal cortex. Inadequate closure manifests as an inability to identify the object until the physical completeness of the drawing approaches an intact contour (Levels 4 or 5).

2. Implicit Perceptual Repetition Priming

Perceptual priming reflects a facilitation in the processing of a physical stimulus as a consequence of prior encounter with that same stimulus, independent of conscious awareness. In the Gollin Figures Test, priming is operationalized as a statistically significant drop in the mean identification threshold between Trial 1 (initial naive exposure) and Trial 2 (retention retest). For example, if a patient initially identifies a horse only at Level 4, but upon re-testing 24 hours later identifies the horse at Level 2, the psychometric shift represents two full stages of perceptual savings. Crucially, this savings effect does not depend on conceptual, semantic, or episodic tagging; it reflects neocortical fluency and structural tuning within extrastriate perceptual representation networks.

3. Explicit Episodic Recognition vs. Implicit Savings

The dual nature of the construct requires distinguishing explicit recollection from implicit perceptual fluency. Explicit memory requires the subject to state: “I have seen this drawing before in this room.” In contrast, implicit memory on the Gollin Figures Test merely requires the perceptual system to resolve the ambiguity of the degraded image faster or with less sensory data. The dissociation between these two constructs is demonstrated when amnesic subjects exhibit normal learning curves on the incomplete figures task while simultaneously performing at chance levels on forced-choice episodic recognition tests evaluating whether the stimulus was encountered previously.

Theoretical Framework

The theoretical foundation of the Gollin Figures Test rests upon two seminal pillars of modern psychological science: Gestalt Perceptual Theory and the Multiple Memory Systems Framework.

Gestalt Psychology and the Law of Closure

Early 20th-century Gestalt theorists, including Max Wertheimer, Wolfgang Köhler, and Kurt Koffka, formulated the fundamental principles of perceptual grouping. Chief among these was the Law of Closure (Prägnanz), which posits that the human mind naturally organizes sensory stimuli into closed, symmetrical, and complete figures rather than fragmented components. Eugene S. Gollin directly operationalized this Gestalt principle by systematically dismantling visual closure through controlled contour deletion. Gollin demonstrated that the threshold at which closure occurs is not purely a physical property of the image, but an active, developmental, cognitive computation governed by central perceptual synthesis and prior experiential tuning.

The Multiple Memory Systems Framework (Schacter & Tulving)

In modern cognitive neuropsychology, the theoretical interpretation of the Gollin Figures Test is governed by the structural memory taxonomy established by Endel Tulving, Daniel Schacter, and Larry Squire. Schacter and Tulving (1994) conceptualized the Perceptual Representation System (PRS) as a pre-semantic, domain-specific subsystem dedicated to the processing of the form and structure of visual words and objects. The PRS is structurally and anatomically distinct from both the Episodic Memory System (which encodes temporal-spatial biographical context via the hippocampus and diencephalon) and the Semantic Memory System (which stores general conceptual knowledge via the anterior temporal lobes).

According to this theoretical model, repetition priming on the Gollin Figures Test is mediated entirely within the structural description subsystem of the PRS, localized within the bilateral lateral occipital and ventral temporal cortices. Functional neuroimaging models (such as fMRI and PET) explain this savings effect through the mechanism of repetition suppression or neural sharpening: upon re-exposure to the incomplete figure, visual processing regions exhibit reduced, more efficient metabolic activity, reflecting the fact that neural ensembles have already optimized their firing patterns to represent the object’s specific spatial geometry.

Validity

Empirical evidence establishing the psychometric validity of the Gollin Figures Test spans more than six decades of neuroscientific, clinical, and experimental research.

Construct Validity

Construct validity has been extensively demonstrated through experimental manipulations of stimulus degradation, presentation intervals, and cognitive loading. Research shows that savings on the Gollin Figures Test are stimulus-specific (perceptual) rather than conceptually generalized. When participants are primed with the written verbal name of an object (e.g., the word “UMBRELLA”) or an acoustically presented spoken word, repetition priming effects on subsequent Gollin fragmented picture identification are markedly attenuated or absent compared to when they are primed with the actual pictorial drawing. This confirms that the test assesses a distinct visual perceptual representation rather than abstract semantic activation.

Criterion and Neuropsychological Dissociative Validity

The criterion validity of the Gollin Figures Test is underscored by its ability to identify double dissociations between amnesic syndromes and visual agnosias:

• Medial Temporal Amnesia: Classic investigations by Warrington and Weiskrantz (1968, 1970) demonstrated that severely amnesic patients suffering from Korsakoff’s syndrome showed significant day-to-day error score reductions on the Gollin figures, matching the performance of age-matched healthy controls, despite demonstrating complete lack of conscious recall. Milner, Corkin, and Teuber (1968) demonstrated that the amnesic patient H.M. showed retention of Gollin figures across a 1-hour to several-day interval.
• Focal Cortical Lesions: Patients with lesions confined to the right temporal-occipital or right parietal-occipital cortices exhibit profound deficits in baseline identification thresholds on the Gollin figures (requiring Level 4 or 5 for initial recognition), confirming the tool’s localization sensitivity to right hemisphere visuospatial integration networks.

Convergent and Discriminant Validity

The Gollin Figures Test exhibits strong convergent correlations ($r = .58$ to $.74$) with other established tests of visual closure and perceptual completion, such as the Visual Object and Space Perception Battery (VOSP) Incomplete Letters subtest, the Hooper Visual Organization Test (HVOT), and Street’s Gestalt Completion Test. Conversely, it shows strong discriminant validity against tests of conscious episodic declarative recall (such as the Rey Auditory Verbal Learning Test [RAVLT] and the Wechsler Memory Scale [WMS]), exhibiting near-zero correlations ($r < .15$) with episodic recall metrics in clinical amnesic populations.

Reliability

The psychometric evaluation of reliability in perceptual threshold and priming paradigms presents unique methodological requirements, as repeated testing inherently contaminates naive baseline performance. Nevertheless, multiple empirical investigations have documented the instrument’s stability and consistency.

Internal Consistency and Split-Half Reliability

Across standard 11-item batteries, split-half reliability coefficients (calculated by correlating identification thresholds across odd and even item sets using the Spearman-Brown prophecy formula) typically range between $r = .72$ and $.86$ across healthy adult cohorts. When computerized expansions of the Gollin paradigm are utilized (e.g., Snodgrass and Vanderwart standard line drawing sets expanded to 20 or 30 incomplete stimulus series), Cronbach’s alpha coefficients for baseline identification thresholds routinely reach $lpha = .81$ to $.89$, indicating strong homogeneity of item difficulty gradations.

Test-Retest Stability and Mnemonic Decay Curves

Assessing test-retest reliability requires separating the psychometric stability of the perceptual threshold metric from the natural decaying trajectory of the implicit memory trace. When alternate, non-overlapping stimulus sets are administered across short intervals to evaluate baseline closure threshold stability, test-retest correlations remain high ($r = .78$ to $.84$).

When assessing the retest savings index across time, empirical studies reveal that while savings scores decline logarithmically as the inter-session retention interval increases (e.g., from 1 hour to 1 month), the rank-order stability of individual priming magnitude among healthy individuals remains moderately stable ($r = .55$ to $.68$). This confirms that individual differences in perceptual repetition fluency reflect a stable psychophysical trait.

Factor Analysis

Structural equation modeling, exploratory factor analysis (EFA), and confirmatory factor analysis (CFA) across diverse clinical and non-clinical cohorts consistently support a two-factor dimensional architecture for the Gollin Figures paradigm:

Factor 1: Perceptual Closure and Structural Decoding

The primary factor accounts for approximately 42% to 54% of the shared variance across items. This factor captures the sensory-perceptual and spatial-integration processing required to assemble fragmented contours into structural descriptions during initial exposure (Trial 1). High factor loadings on this dimension ($lambda > .60$) are observed across complex objects with intricate interior lines and curved contours (e.g., animals, intricate mechanical objects). Performance loading on Factor 1 is strongly correlated with non-verbal fluid intelligence, visuospatial processing indices, and right parieto-occipital integrity.

Factor 2: Perceptual Repetition Priming / Savings Magnitude

The second distinct factor accounts for 20% to 28% of the total variance and is defined by the delta scores across trials (Trial 1 threshold minus Trial 2 threshold). CFA models indicate that Factor 2 operates orthogonally to episodic memory factors (e.g., verbal delayed recall, paired-associate learning). Structural equation models confirm that while Factor 1 is significantly influenced by age and general cognitive speed, Factor 2 (implicit perceptual priming) exhibits remarkable resilience across healthy adult aging, showing minimal decay across cohorts aged 20 to 80 years.

Instrument / Measurement Tool

The Gollin Figures Test is an individually administered performance-based psychometric instrument. While originally developed as a physical booklet/card battery, modern administrations encompass standardized paper protocols, slide projection formats, and computerized software displays.

• Test Type: Performance-based visual perceptual and non-declarative memory test; psychophysical ascending threshold paradigm.
• Format: Visual stimulus presentation using an ascending method of limits. Cards or digital screens display line drawings degraded across five standardized levels.
• Item Count: The classic Gollin battery comprises 11 distinct object series (e.g., telephone, fish, umbrella, automobile, table, shoe, etc.). Each series contains 5 graded cards, yielding a total of 55 standardized stimulus cards.
• Stimulus Gradations:
  • Level 1: Maximum degradation (~80% to 90% contour deletion; sparse, disjointed lines).
  • Level 2: Heavy degradation (~65% to 75% contour deletion).
  • Level 3: Moderate degradation (~50% contour deletion; major structural outline emergent).
  • Level 4: Minimal degradation (~25% contour deletion; near-complete drawing).
  • Level 5: Intact canonical contour drawing (100% complete line illustration).
• Administration Procedure:
  • Trial 1 (Baseline Acquisition): The examiner presents the drawings of an item series in strictly ascending sequence, beginning with Level 1. Each card is displayed for a standardized duration (typically 2 to 3 seconds). The respondent is instructed to state what the drawing depicts as soon as it is recognized. If identification is incorrect or absent, the examiner proceeds immediately to Level 2, and so forth, until the respondent correctly names the object. The grade level at which correct recognition occurs is recorded as the item threshold (ranging from 1 to 5). This process is repeated across all 11 item series.
  • Retention Interval: An experimental delay is introduced. Depending on the clinical or investigative protocol, the retention interval can be immediate, short-term (10 to 60 minutes), or long-term (24 hours to multiple days/weeks).
  • Trial 2 (Retention/Priming Evaluation): The 11 series are presented again using the identical ascending method of limits. The new identification threshold for each item is recorded.
• Scoring Metrics:
  • Total Error Score: The sum of errors across all 11 items. If an object is recognized at Level 1, zero errors are scored. If recognized at Level 3, two errors are scored (Levels 1 and 2 failed). The maximum possible error score per series is 4 (if recognized only at Level 5). The overall test error score ranges from 0 to 44.
  • Mean Recognition Threshold: The average card level required for correct identification across the 11 items: $\bar{T} = \frac{\sum \text{Thresholds}}{11}$ (ranging from 1.0 to 5.0).
  • Savings Score (Absolute Priming Magnitude): Computed as: $\text{Savings} = \text{Error}_{\text{Trial 1}} – \text{Error}_{\text{Trial 2}}$, or alternatively: $\Delta \bar{T} = \bar{T}_{\text{Trial 1}} – \bar{T}_{\text{Trial 2}}$.
  • Proportional Savings Percentage: Expressed as the proportion of initial baseline error eliminated during the retest: $% \text{Savings} = \left( \frac{\text{Error}_{\text{Trial 1}} – \text{Error}_{\text{Trial 2}}}{\text{Error}_{\text{Trial 1}}} \right) \times 100$.

Permissions & Fee and Test Year

The Gollin Figures Test was first published by Eugene S. Gollin in 1960. Because the original stimulus set was published in academic literature prior to modern commercial proprietary software conventions, the foundational theoretical paradigm and the original 11 line drawing series reside within the academic public domain. Various standardized card sets, computerized versions, and commercial distributions have subsequently been published by neuropsychological test distributors, university laboratories, and psychological research consortia.

Researchers and clinicians intending to administer the classic Gollin Figures Test or its modern standardized variants (such as the Snodgrass & Vanderwart fragmented image sets or digital psychometric adaptations) can generally access stimulus specifications, line drawings, and normative lookup tables directly through peer-reviewed research publications, institutional clinical repositories, or academic neuropsychology handbooks without prohibitive licensing fees. However, digitized testing software, computerized touch-screen systems, and formalized commercial clinical kits bearing proprietary normative datasets may require software licensing or clinical kit purchase through their respective distributors.

References

The following peer-reviewed publications document the empirical development, clinical validation, and neuropsychological applications of the Gollin Figures Test:

• Corkin, S. (1968). Acquisition of motor skill after bilateral medial temporal-lobe excision. Neuropsychologia, 6(3), 255–265. https://doi.org/10.1016/0028-3932(68)90024-9
• Gollin, E. S. (1960). Preparing visual materials for developmental study: The incomplete figures test. Perceptual and Motor Skills, 11(1), 93–98. https://doi.org/10.2466/pms.1960.11.1.93
• Gollin, E. S. (1961). Further studies of visual transfer of training in developmental study: The incomplete figures test. Perceptual and Motor Skills, 13(3), 307–312. https://doi.org/10.2466/pms.1961.13.3.307
• Gollin, E. S. (1962). Factors affecting the visual recognition of incomplete images: A comparative study. Perceptual and Motor Skills, 15(3), 583–590. https://doi.org/10.2466/pms.1962.15.3.583
• Gollin, E. S. (1965). A developmental approach to learning and cognition. In L. P. Lipsitt & C. C. Spiker (Eds.), Advances in Child Development and Behavior (Vol. 2, pp. 159–186). Academic Press. https://doi.org/10.1016/S0065-3233(08)60312-3
• Milner, B., Corkin, S., & Teuber, H.-L. (1968). Further analysis of the hippocampal amnesic syndrome: 14-year follow-up study of H.M. Neuropsychologia, 6(3), 215–234. https://doi.org/10.1016/0028-3932(68)90021-3
• Schacter, D. L. (1987). Implicit memory: History and current status. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13(3), 501–518. https://doi.org/10.1037/0278-7393.13.3.501
• Schacter, D. L., & Tulving, E. (1994). Memory Systems 1994. MIT Press.
• Snodgrass, J. G., Smith, B., Feenan, K., & Corwin, J. (1987). Identifying degraded visual displays: Systems, thresholds, and strategies. Journal of Experimental Psychology: General, 116(3), 240–268. https://doi.org/10.1037/0096-3445.116.3.240
• Squire, L. R. (1992). Memory and the hippocampus: A synthesis from findings with monkeys and humans. Psychological Review, 99(2), 195–231. https://doi.org/10.1037/0033-295X.99.2.195
• Tulving, E., & Schacter, D. L. (1990). Priming and human memory systems. Science, 247(4940), 301–306. https://doi.org/10.1126/science.2296719
• Vokey, J. R., Baker, S. W., Hayman, G., & Jacoby, L. L. (1986). Perceptual identification of visual patterns. Perception & Psychophysics, 40(6), 422–426. https://doi.org/10.3758/BF03208200
• Warrington, E. K., & Weiskrantz, L. (1968). New method of testing long-term retention with special reference to amnesic patients. Nature, 217(5132), 972–974. https://doi.org/10.1038/217972a0
• Warrington, E. K., & Weiskrantz, L. (1970). Amnesic syndrome: Consolidation or retrieval? Nature, 228(5272), 628–630. https://doi.org/10.1038/228628a0

Items of the Scale