UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Completion of occluded objects in early vision : an exploration of spatial limits Shore, David I. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_fall_shore_david.pdf [ 2.62MB ]
JSON: 831-1.0086297.json
JSON-LD: 831-1.0086297-ld.json
RDF/XML (Pretty): 831-1.0086297-rdf.xml
RDF/JSON: 831-1.0086297-rdf.json
Turtle: 831-1.0086297-turtle.txt
N-Triples: 831-1.0086297-rdf-ntriples.txt
Original Record: 831-1.0086297-source.json
Full Text

Full Text

Completion of Occluded Objects In Early Vision: An Exploration of Spatial Limits by  David I. Shore B.Sc., McMaster University, 1991 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF PSYCHOLOGY We accept this thesis as conforming to the required standard Dr. J.T. Enns Dr. L.M. Ward Dr. C. Blaha  THE UNIVERSITY OF BRITISH COLUMBIA August, 1993 © DAVID I. SHORE, 1993  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department of  25 1 C' L€( 5Y (  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  ocf- 1 ‘1-, (4 q3  Abstract Our visual experience is of complete objects despite the fact that the retina is often given only partial views of these objects. Objective support for this perceptual completion of partly occluded objects was generated in Experiment 1 in which same-different response time (RT) was measured for objects that were either complete, partly occluded, or notched.^Complete and partly occluded objects yielded similar RTs, while notched objects took much longer to match. For larger amounts of overlap the occluded condition took longer than the whole condition. This suggested that rapid completion is limited to relatively small gap sizes. In Experiment 2, we varied the amount of occlusion systematically in order to find the limit of rapid completion. With 25% object occlusion or less, we again found evidence for rapid completion; beyond this point subjects appeared to use a strategy that was similar to that used in the notch condition. The separability of completion from other cognitive processes was investigated using two strategies. First, in Experiment 3 the task demands were varied from an identification task to a categorization task. ^Similar spatial limitations were found for both of these tasks. Second, in Experiment 4 the target was a moving object and it was again found that completion was separable from other later processes such that moving targets were easier to complete but no easier to classify. These results support the proposal of a preattentive completion process that is spatially limited.  ^  Table of contents Abstract  ii  Table of contents  ^iii  List of Tables  ^v  List of figures  ^vi  Acknowledgments ^ Introduction^  vii 1  Experiment 1 ^ Method ^ Subjects ^ Stimuli Apparatus ^ Procedure ^ Results ^ Discussion ^  13 16 16 16 17 18 20 24  Experiment 2a ^ Method ^ Subjects ^ Stimuli Apparatus ^ Procedure ^ Results ^ Discussion  25 27 27 27 28 28 29 31  Experiment 2b ^ Method Subjects ^ Stimuli Apparatus ^ Procedure. Results ^  32 32 32 33 33 33 33  ^  36  Discussion ^ Experiment 3 ^ Method ^ Subjects ^ Stimuli Apparatus ^ Procedure ^ Results ^ Discussion ^  38 40 40 40 40 40 41 44  Experiment 4 ^ Method ^ Subjects ^ Stimuli Apparatus ^ Procedure ^ Results ^ Discussion ^  45 46 46 48 49 49 49 54  General Discussion ^  55  References ^  63  Appendix A - Raw Tabled Data ^  66  iv  List of Tables Table 1. Results of Experiment 1: Raw Data (RT and Error) ^ 6 7 Table 2. Results of Experiment 2a: Raw Data (RT and Error) ^ 68 Table 3. Results of Experiment 2b: Raw Data (RT and Error) ^ 6 9 Table 4. Results of Experiment 3: Raw Data (RT and Error) ^ 70 Table 5. Results of Experiment 4 - Static Conditions: Raw Data (RT and Error) ^  71  Table 6. Results of Experiment 4 - All Move Conditions: Raw Data (RT and Error) ^  72  Table 7. Results of Experiment 4 - Target Only Move Conditions: Raw Data (RT and Error) ^  73  List of figures Figure 1 Phenomenal Demonstration of Early and Late Completion^3 Figure 2. Sample Stimuli and Naming System  ^5  Figure 3. Example of Line Fragmentation Task (Enns & Rensink, in press) 7 Figure 4. Example of Notched Task (Enns & Rensink, in press) ^ 10 Figure 5. Sample Stimuli in Experiment 1 ^  14  Figure 6. Results in Experiment 1 ^  22  Figure 7. Results In Experiment 2a ^  30  Figure 8. Results in Experiment 2b ^  35  Figure 9. Results in Experiment 3 ^  42  Figure 10. Sample stimuli used in Experiment 4 ^  47  Figure 11. Results in Experiment 4 ^  53  vi  Acknowledgments  First and foremost, I would like to thank Dr. Enns for his support, both mental and material, in this project. From the very beginning he made me feel comfortable and welcome. As well, I would like to thank my committee members for their suggestions and patience; especially Lawrence Ward for his seemingly unending knowledge base and willingness to share. To the Enns' lab members who have made my work environment a happy and conducive one I extend my warmest thanks: Deb Aks for her mentorship in the ways of being a Grad student (what a good influence you were); Darlene Brodeur for her Tye Dye Parties and the patient way she answers all my questions; Lana Trick for her suggestions, and all the coffee we drank; Tracy Wood, for all her help in getting things done, running some of my subjects and for just being around to brighten up the lab. I also wanted to recognize the contribution of Geoff Small for his participation in running the first experiment. On a more personal note I owe perhaps my greatest thanks to Trudy Mohrhardt for her endless devotion and constant support. Other friends that need mention include Matt Pollack, Jon and Devin Summerhayes for their editing suggestions; Zonya, Matt, Wilf and Gaia for allowing Trudy and I to live with them for so long; and David Tessler who has introduced me to more B.C. wonders than I can remember. Finally I want to thank my family members who have supported me throughout my education; Barb, Abe, Pearl, and Elan. And a special thanks to Dave Shore (the other one) for introducing me to all of my Vancouver Friends.  vii  Early Completion 1 Introduction We experience the visual world as a collection of whole objects despite the fact that the retina receives only fragmentary information about them. For instance, when distant objects are partly occluded by nearer objects we nevertheless perceive the more distant objects as whole. This phenomenological observation leads to questions such as: "How does the visual system go beyond the information given to generate the experience of a whole object?", "How early in the visual stream is this interpolation conducted?" and, "Are there limitations in the temporal and spatial extent of this proposed process?". In this study I will show that there are spatial limits to the completion process. These limitations follow logically from the claim that object completion is an early vision mechanism (Enns, 1992; Zucker, 1987). Kanizsa (1979) describes two perceptual processes we use to go beyond the fragmentary information on the retina. The primary one "consists of an organization of proximal stimulation (Kanizsa, 1979, pp5)" which includes the completion of partly occluded objects using simple gestalt rules. The secondary process acts on the information supplied by the first to make perceptual inferences and behavioral decisions. The validity of this distinction can be observed when  Early Completion 2 comparing Figure 1 a and 1 b, which are both representations of the Necker cube (Figure 1c). In Figure 1 a it is very difficult to perceive the entire cube and even more difficult to experience the depth reversal for which the Necker cube is famous. If one is able to identify Figure 1 a as a Necker cube it is presumably because the presented fragments can be integrated into a somewhat complete whole based on past experience. In Figure 1 b there is no requirement for such inferences because the object is completed by a primary process and presented to consciousness as a complete Necker Cube (Figure lc). Thus, the visual system can be dichotomized into two processes: a primary one which can organize the proximal stimulus into larger wholes following Gestalt like rules, and a secondary one that can reconstruct and recognize objects based on previous knowledge and the information passed on from the primary process. This information may, at times, be nothing more than the fragmentary information presented on the retina. In proposing this primary/secondary dichotomy, researchers generally assume that primary completion processes are active early in the visual stream by a fast, automatic and spatially parallel mechanism (Enns, 1992; Enns & Rensink, 1993; Enns & Rensink, in press; Kanizsa & Gerbino, 1982; Kellman & Shipley, 1992; Ramachandran, 1992, 1993).  Early Completion 3  A  B  C Figure 1. Phenomenal Demonstration of Early and  Late Completion  Early Completion 4 Much of the previous research addressing the question of object completion has used phenomenological methodologies or appeals to subjective impressions (Kanizsa, 1979; Kellman & Shipley, 1992). More recent efforts, including the current work, have used performance based measures to study the completion mechanism in an objective fashion (Enns, 1992; Enns & Rensink, in press; Gerbino & Salmaso, 1987; Sekular & Gaber, 1993; Sekular & Palmer, 1992). Before describing these procedures and findings, I will discuss the naming of stimuli, since no consistent naming practices can be found in the literature. For the most part, discussion will be limited to the first three conditions shown in Figure 2. In the whole condition (Figure 2a) a target item is presented in its entirety either on top of another object or by itself. In the occluded condition (Figure 2b), the target object is presented behind an irrelevant occluder. In the notched condition (Figure 2c), the target object has the same outline as the occluded object with the exception that there is no visible occluder. Thus a retinal difference between the whole and occluded objects is the occluded region of the occluded object. Similarly, an important difference between the notched and occluded objects is the presence of the occluder in the notch for the occluded condition.  Early Completion 5 Condition Name  Experiment 1 (Gerbino and Salmaso, 1987)  Experiments 2, 3, and 4  A) whole (TPC) complete  B) occluded (PC)  C) notched (C) incomplete  D) occludedtruncated (PC.t)  E) wholetruncated  N/A  Figure 2. Sample stimuli and naming system  Early Completion 6 The remaining conditions are used to control for several possible confounds. For example, reaction time differences between the notched and occluded objects may be due to the presence of the concavity on the notched object (See Figure 2). To control for this interpretation, occluded and whole conditions are often run with notched distractors so that the same junction information is present. The occluded-truncated (Figure 2d) and the whole-truncated (Figure 2e) conditions are examples of this type of control figure. For the remainder of this paper I will make exclusive use of these terms. Several experiments have been conducted to investigate the ability of early vision to complete objects behind perceived occluders. Enns and Rensink (in press; Enns, 1992) have looked at two types of occlusion situations: one involving the completion of simple line segments across a filled or empty gap, and another looking at the completion of 2-D geometrical shapes that have a portion missing. In the first case the two fragments can be considered as separate entities or as part of a common object. In the second case there is only one fragment that simply has a portion missing. Both of these lines of research have used the visual search task (Treisman & Galade, 1980).  Early Completion 7  Figure 3. Example of line fragmentation task  (Enns & Rensink, in press).  Early Completion 8 In the line fragmentation task, the attribute that distinguished the target from the distractor was the location of the gap, either in the centre of the line segment (distractors) or offset to either side (target, See Figure 3). The subjects task was to indicate if a target item was present or not. When the gaps were unfilled (Figure 3a) the task was very easy (search slopes averaged 6 ms per item for target present and 4 ms per item for target absent) as subjects could use the length of the visible fragments as features on which to base their search. When the gaps were filled with a 2-d occluder (figure 3b) the task became more difficult (29 ms and 55 ms per item for target present and absent respectively), presumably because subjects were completing behind the objects and thus the target and distractors did not differ in overall length. When a 3-d occluder was present in the gaps (Figure 3c) the task became even more difficult (47 ms and 88 ms per item) indicating that completion was even more efficient. The fact that search was difficult when an occluder was present but easy when no occluder was present supports the claim that completion behind an occluder occurs preattentively. If subjects did not complete both the target and distractor line segments very rapidly and in parallel then the results for the second two experiments would look more like the first experiment. Finding steeper search slopes indicates that  Early Completion 9 completion is occurring before a parallel search can be conducted. Note that this experiment and the following one directly support the proposal that completion occurs automatically (i.e. without voluntary control) because subjects complete even though it is to their advantage not to. The second set of figures used by Enns and Rensink (in press) is more similar to the stimuli used in the current investigation (Figure 4). In this task, the subject searched for a notched black square. The distractors were whole black squares. When the notch was empty (Figure 4a) and could thus serve as a feature for which to search, the task was very easy (7 ms and 8 ms per item). On the other hand, when the gap was filled by a visible occluder (Figure 4b), search was very difficult (36 ms and 66 ms per item). The retinal stimulation from the targets in these two situations was identical. The difference in reaction time slopes between the conditions can be explained by the automatic completion of the occluded gap such that the representation of the target object was the same as the distractor items (whole squares) and thus required attention to locate. This experiment again demonstrates that completion occurs before a parallel search can be conducted (i.e. early in vision).  Early Completion 10  A  B  Figure 4. Example of notched task (Enns & Rensink, in press).  Early Completion 11 The rapid nature of the completion process was investigated by Sekular and Palmer (1992) using a high speed priming paradigm. They primed subjects with a whole, occluded, or notched circle or square, then required subjects to make a same-different response to a pair of complete or incomplete items in the probe. The stimulus onset asyncrony (SOA) between the prime and probe was varied systematically between 10 ms and 400ms. When the prime was a whole or notched figure, subjects were faster at matching the corresponding complete or incomplete figures regardless of SOA. When the prime was an occluded figure there was an interesting interaction with SOA. Before an SOA of 200 ms, subjects were faster to match a pair of incomplete items whereas after 200 ms they were faster on the complete items. This indicates that before 200 ms, subjects still had a fragmented representation of the occluded figure, while after 200ms this representation was completed. One final experiment in the literature speaks to the rapid nature of completion and follows more directly from the phenomenal demonstration presented earlier in the introduction. Gerbino and Salmaso (1987) had subjects match whole shapes with comparison shapes that were either whole, occluded, or notched. They found that reaction times to match the occluded and whole shapes did not differ whereas reaction times for the  Early Completion 12 notched condition were significantly longer. This was taken to indicate that the notched shapes were being matched by a different, slower and later process, perhaps one corresponding to Kanizsa's secondary process. The occluded objects were completed using the primary process and then matched to the whole object. Presumably all of these objects are matched using some attention-based operation and the advantage found for the whole and occluded objects reflects an easier feature matching process than that needed for the notched figures. In summary, it seems that partly occluded objects can be completed rapidly, automatically, and spatially in parallel.^These characteristics define the completion of occluded objects as a preattentive process. One implication of the spatially parallel nature of completion is the existence of local operators of limited spatial extent (Enns, 1992; Enns & Rensink, 1993; Zucker, 1987). For completion to be conducted across the scene in parallel each local area must be processed independently, implying that the proposed operator must have a limited neighbourhood of operation. When the area of occlusion does not fall within this area, early completion should not be successful and the fragmented version presented to the retina would be passed on to later visual routines for identification. The purpose of this study is to confirm that the proposed  Early Completion 13 local operator is indeed spatially limited. ^It was predicted that for larger gap sizes than those sampled in the literature, there would be a failure of the completion process within early vision. To test this prediction, a replication of the Gerbino and Salmaso (1987) experiment was conducted for two gap sizes. Experiment 1 The first experiment had two purposes: to replicate the Gerbino and Salmaso (1987) result using our current apparatus, and to determine if the results associated with occluded objects having small degrees of overlap would generalize to objects with larger gap sizes. There were two reasons for examining the larger gap sizes. First, all of the previous experiments in the literature used objects that were only occluded to a very small extent (i.e. at least 75% of the object was visible) and this limitation has theoretical implications for the completion process. Second, just as Sekular and Palmer (1992) showed that the proposed completion process is temporally limited, we need to discern if it is also spatially limited, as is predicted by the characteristics of preattentive processes. To these ends, the Gerbino and Salmaso (1987) study was replicated for two levels of occlusion. The small gap size was comparable to the original stimuli, whereas the large gap stimuli had about one third  Early Completion 14  Condition Name  A)whole (TPC) complete B)occluded (PC) C) notched (C) incomplete  Small Gap Size  4WAAlk Alk,  Large Gap Size  Abal  AA,  otAik  Figure 5. Sample stimuli for Experiment 1  Early Completion 15 more of the object covered. Examples of the stimuli used can be seen in Figure 5. It was predicted that for the small gap size, like in the Gerbino and Salmaso (1987) study, the whole and occluded stimuli should not differ in terms of reaction time or accuracy of matching. On the other hand, the notched condition should show longer reaction times and decreased accuracy in comparison to both the whole and the occluded conditions. For the large gap size, several patterns of results were possible depending on the nature of the completion process. If completion is general and can complete any gap size quickly then the same pattern of results might be expected. Alternatively, if completion is spatially limited, the occluded condition would be more difficult to match than the whole condition for the large gap size only if indeed the large gap exceeded the size of the local operator field. There are two operational definitions that will be used to show that subjects are using the early vision strategy associated with the completion process. The first concerns the difference, in terms of RT and accuracy, between the whole condition and the occluded condition. If completion occurs early in processing, then these two stimuli should be perceived as the same and not show any differences in reaction time. The  Early Completion 16 second operational definition pertains to the notched condition. Since the retinal stimulation, in terms of the target item, is identical in occluded and notched conditions, the subjects' RT and accuracy should also show similar patterns. To the extent that the occluded condition is faster than the notched condition we can infer that early object completion has occurred. Method Subjects. Ten volunteer subjects were drawn from the undergraduate subject pool at the University of British Columbia. In return for participating, subjects received extra credit in a psychology undergraduate course. There were five right handed males, four right handed females and one left handed female. All subjects had normal or corrected to normal vision. Stimuli. A display consisted of a single target item displayed 7° above or below fixation and a comparison configuration consisting of either one or two items displayed opposite the target item (also 7° above or below fixation). Individual items consisted of diamonds (2.9° x 2.9°), triangles (3.8°x1.9°), and hexagons (3.5°x1.8°). The comparison items could be spaced at one of three distances. In the small gap condition, the overlap was about 0.3° of visual angle (comparable to the stimuli used in  Early Completion 17 Gerbino & Salmaso, 1987). For the large gap, this value was about 1.2° of visual angle. In the control condition, there was no overlap and there was either one or two items (0.25° apart). The target could have one of the five relations to the comparison item. The first four possible combinations were given in the introduction under naming systems (See Figure 2 and page 4). In addition to these four conditions (whole, occluded, occluded-t, notched) 50% of the trials required a different response. In these trials the comparison items were different from the target item. In total there were fourteen stimulus configurations. Ten corresponded to the cells in a two by five factorial design where the two factors were gap size (small and large) and condition (whole, occluded-c, occluded-t, notched, and different). The remaining four trial types were control conditions. There were two types of control conditions: single and double. In the single condition there was only one comparison item which was either the same as or different from the target item. In the double condition there were two comparison items. These items could either both be different than the target or one of them could be the same as the target. See Figure 5 for examples of the whole, occluded, and notched conditions.  Early Completion 18 Apparatus. Subjects responded using a standard Macintosh (Mac) keyboard which was connected to a Mac plus. The stimuli were presented on the Mac plus screen which had the contrast and brightness set to the highest level. VScope software (Rensink & Enns, 1992) was used to generate the stimuli and record millisecond reaction times and error rates. Procedure. When subjects arrived they were seated in front of a computer and asked to fill out two forms. The first was a simple sign-in sheet. The second consisted of an instruction sheet with consent form attached and a biographical questionnaire asking about age, handedness, sex, eyesight, and year of university. Subjects were told to read the instructions carefully and sign the consent form if they agreed to participate. The instructions emphasized equally the importance of accuracy and speed of responding. After the subjects signed the consent form they were asked if they had any questions. Following these introductory procedures the experimenter emphasized the important points of the task. First, subjects were to try to keep their eyes fixated on the centre of the screen throughout the experiment. Second, the fact that some of the items would be truncated was discussed and it was  Early Completion 19 emphasized that these items should be treated as exemplars of the complete form. Third, the response keys to be used were reviewed. Subjects were then given at least 15 trials of practice. They practiced until they could perform 5 consecutive trials correctly. If the subject was having difficulty, the instructions were reviewed again and they were prompted for any questions. Subjects sat with their faces approximately 40 cm from the screen and were free to move. The task was to indicate if one of the comparison items was the same as the target item. They used the index finger of one hand on the keyboard to indicate same and the index finger of the other hand to indicate different. The side of responding was counterbalanced across subjects. Each block began by prompting the subject to press any key to begin. When subjects signalled they were ready, the first display appeared after 750ms. The display lasted for 200ms. After the subject responded or after three seconds had elapsed a feedback symbol was presented for 915ms. This served as the fixation for the next trial which was presented after a 750ms inter-trial-interval during which the screen was blank. Feedback consisted of either a plus sign, a minus sign, or an open circle indicating that the preceding response was either correct,  Early Completion 20 incorrect, or beyond the three second limit. As well, at the end of each block the total percent error for that block was presented to the subject. There were 10 blocks of 60 trials each. The subject was encouraged to take self-paced breaks between blocks. The experimenter remained in the room during the practice trials to answer any questions and monitor for the criterion number of correct responses. During the actual experiment the experimenter left the room but was available if the subject had questions or problems. Results The first block of trials and the first three trials of every subsequent block were treated as practice and not included in the analysis. This left 504 trials to be analyzed for each subject. For these trials the error rate was calculated for each condition by dividing the number of incorrect trials by the total number of trials for that condition. The mean reaction times for the remaining correct trials was also calculated for each subject. These mean reaction times and error rates were submitted to a repeated measures ANOVA. Follow up analyses and a priori predictions were conducted using either a simple main effects ANOVA for  Early Completion 21 interactions or the Fisher LSD procedure for pairwise comparisons. The MS e used was always taken from the highest order interaction of the overall analysis. Sphericity assumptions were tested using both the Greenhouse & Geiser (1959) technique and the Huynh & Felt (1970) technique. Throughout the four experiments there were no serious violations of the sphericity assumption. Whenever there was a violation, the lower probability value will be presented. The same trials were submitted to a two by four repeated measures ANOVA where the two factors were aap size (small and large) and condition (whole, occ, occ-t, notch). The main effect of gap size was significant [F(1,9)=5.3, MS e =6909.7, p<.05] as was the effect of condition [E(3, 27 )=35.9, MS e =6212.5, p<.0001]. These differences were moderated by a significant interaction [F(3,27)=10.9, MS e =2486.7, p<.0001]. As can be seen in Figure 6, this interaction was due primarily to the significant difference between the occluded trials and the whole trials for the large gap size [t(27)=-5.9, p<.0001], but not for the small gap size [t(27)=-0.7]. The occluded trials were faster in the small gap condition than in the large gap condition [t(27)=-4.7, p<.0001]. The same was true for the notched trials [t(27)=-4.8, p<.0001]. As well, the occluded trials were generally faster than the notch trials [t(27)=-6.1, p<.0001].  Early completion 22  Figure 6. Results of Experiment 1  Early Completion 23 In terms of errors, the main effect of condition was significant [F(3,27)=25.5, MS e =.012, p<.0001] while that for gap size was not [E(1,9)=4.3, MS e =0.011]. These differences were again moderated by a significant interaction [F(3,27)=4.1, MS e =.009, p<.05]. Although the errors follow the same pattern as the reaction times, the only significant differences were found in the notch conditions which were more error prone than any other condition [t(27)=9.54, p<.0001], with the large gap size being more error prone than the small gap size [t(27)=-4.1, p<.0005]. You may note that the occluded-truncated trials did not differ from the whole trials (See Table 1 for raw data) for the small or large gap sizes in terms of reaction times [t(27)=-.05 for small and t(27)=.399 for large] or error rates [t(27)=-.19 for small and t(27)=.449 for large] (See Table 1). This was expected for the small gap size but seems counterintuitive for the large gap size, given the large effect found for the occluded condition. This finding probably has to do with the choice of stimulus configurations. Essentially, the occluded-t trails are present as a result of the need for notched trials and the desire to have all possible pairings of items presented throughout the experiment. Thus, the large gap size for these trials refers to the gap in the truncated occluder and not to the amount of overlap between the two items. This result, then,  Early Completion 24 implies that the important variable for these stimuli is the amount of overlap and not the notch size in the occluding distractor. In comparing same and different trials a somewhat surprising speedaccuracy trade off was discovered such that same trials were faster than different trials [E(1,9)=31.7, MS e =4069.5, p<.0003] although more error prone [F(1,9)=5.8, MS e =.002, p<.05]. This trade off was entirely due to the already mentioned high error rates in the notch condition for both the small and large gap sizes.  Discussion For small amounts of overlap, subjects were able to respond to the occluded objects in the same way as to the whole objects. Both of these conditions were significantly faster than the notched condition. Thus, according to both of our operational definitions subjects seem to be using the early completion process for these small gap stimuli. This replicates the Gerbino and Salmaso (1987) finding. For large amounts of overlap, however, the occluded trials took significantly longer than whole trials. The notched condition was always slower and more error prone than either of the occluded or whole conditions. In terms of our operational definitions these results are somewhat ambiguous. Since the occluded trials were faster than the  Early Completion 25 notched trials we must assume that some information from early in the visual stream is aiding the later categorization of the object fragment. However, since the occluded condition was slower than the whole condition there must be some degree of failure of the early completion process. In terms of the two goals, the original study showing that the early vision strategy can operate on slightly occluded objects was replicated. As well, for large amounts of overlap, occluded objects required a longer amount of time and were more prone to errors than whole objects. Thus, the operation of this early strategy does indeed seem to be spatially limited in its extent of operation.  Experiment 2a The purpose of the Experiment 2 was to investigate the spatial limit of early completion in a more systematic way. In Experiment 1 there were only two gap sizes. As well, the information obtained from the large gap condition was equivocal in terms of the two operational definitions. For small gap sizes, completion operated at full efficiency. The other end of the continuum where completion fails entirely was not found. Instead,  Early Completion 26 for the large gap sizes, there was some evidence of completion by one operational definition but not by the second. There were two methodological simplifications from Experiment one which were hypothesised to reduce the overall number of processing stages that subjects must engage. First, the stimulus set size was decreased from three items to two, thus reducing the memory load required to recognize the shapes. Second, the task was changed from matching to a two alternative forced choice identification. This had the advantage of eliminating the need for visual comparisons. The shapes used were circles and squares which could be either black or white on a neutral gray background. The subjects' task was to identify the black object as either a circle or a square. The white item was always irrelevant and to be ignored by the subjects. By reducing the task demands on the subjects, the added variability from other processing stages would be presumably reduced. The gap size measurements are given as the percentage of area that was covered by the occluding object. The four levels tested in this experiment were 25%, 50%, 70%, and 85%. The conditions tested were also slightly altered in this experiment to test several different hypotheses. The occluded-truncated condition was  Early Completion 27 replaced with a whole-truncated condition. The whole-truncated condition was added to see if manipulations of the occluded, irrelevant distractor had any effect on identification of the whole target. The occludedtruncated condition was removed to keep the total number of conditions down. Again there were two control conditions, single and double which had, respectively, one or two non-overlapping items. As well, there was a notched-alone condition which was included to see if the effect of a distractor varied with the size of the gap.  Method Subjects. There were four right handed males and six right handed females. All subjects had normal or corrected to normal vision. Subjects were drawn from the same subject pool as Experiment 1.  Stimuli. A display consisted of a stimulus set displayed 6° to the right or left of fixation^Stimuli consisted of circles (2.2°) and squares (2.0°) which could be either black or white. The background was a neutral gray where 50% of the pixels were black and 50% were white (see figure 2 right hand side). There were twenty two conditions. Twenty of these corresponded to a four by five factorial design where level of occlusion and trial type were the two factors. The levels of occlusion that were tested included 25, 50,  Early Completion 28 70, 85 percent occluded objects. The following trial types were included: whole, occluded, notched, notched-alone, and whole-truncated. The last two conditions, which were not represented in the factorial design, were the control conditions single and double. Note that the notched-alone condition was identical to the notched condition except that no distractor was presented behind the fragment. The black item always appeared on the midline of the screen. If the target was on the right side of the screen the distracting object could appear above and to the right or below and to the left of the target. If the target appeared on the left side of the screen, the distracting object could appear above and to the left or below and to the right of the target. The side of presentation and the location of the distractor were randomly chosen from trial to trial by the computer. Apparatus. The equipment and software used were the same as in  Experiment 1. Procedure. Overall procedures, display durations, and feedback were  identical to Experiment 1. The task was to indicate the identity of the black item. Subjects used the index finger of one hand on the keyboard to indicate circle and the  Early Completion 29 index finger of the other hand to indicate square. The side of responding was counterbalanced across subjects. There were 20 blocks of 50 trials. Results After the first block of trials and the first three trials of each subsequent block were removed there were 893 trials to analyze per subject. Reaction times were submitted to a four by five repeated measures factorial ANOVA. The first factor was level of occlusion and included 25%, 50%, 70%, and 85% occlusion. The second factor was condition and included whole, occluded, notched, notched-alone, wholetruncated. Both main effects were significant as was the interaction of the two [E(4,36)=37.5, MS e =880.5, p<.0001 for type of trial; E(3,27)=31.6, MS e =536.6, p<.0001 for level of occlusion; F(12,108)=6.2, MS e =281.3, p<.0001 for the interaction] (See Figure 7 and Table 2). It had been predicted that for small amounts of overlap the occluded condition would be as fast as the whole condition and much faster than the notched condition. For large gap sizes the reverse was predicted such that the occluded condition would be much slower than the whole condition and not differ from the notched condition. Contrary to our first prediction, the occluded trials were significantly different from the  Early Completion 30  Figure 7. Results of Experiment 2a  Early Completion 31 whole trials at all levels of occlusion including the smallest [t(108)=3.4, p<.0014 for the 25% occlusion condition]. In fact the occluded condition was not different from the notched condition overall [t(36)=.614]. Only at the smallest level of occlusion (25%) was the occluded condition faster than the notched condition [t(108)=2.4, p<.02] as was predicted. Thus we confirmed three of our four predictions. First, for small gap sizes, the occluded condition was faster than the notched condition. Second and third, for large gap sizes, the occluded condition was equal to the notched condition, and much slower than the whole condition. The prediction that for small gap sizes the occluded condition would be as fast as the whole condition was not confirmed. Discussion  For large gap sizes the occluded condition required the same amount of time as the notched condition. Both of these conditions were slower than the whole condition. For the smallest gap size (25%) it was found that the occluded condition was faster than the notched condition but not as fast as the whole condition. Thus the 25% level of occlusion was similar in pattern of results to the large gap size condition in Experiment 1.  Early Completion 32 This experiment has isolated the opposite end of the continuum from the first experiment, that is, the end where there is no information available to later cognitive levels of processing from the early completion process. Experiment 2b was run to sample the levels of occlusion between 3% and 46%. Experiment 2b This experiment was conducted to investigate at what point the completion mechanism fails. The previous experiment showed that by 50% occlusion there is no evidence of completion and that at 25% there was marginal evidence for completion. The levels sampled for this experiment fell between 3% and 46% to give a more fine grained picture of this important region. From the previous experiment we predicted that for occlusion levels greater than 25% there should be no completion while at 25% there should be some completion. At some point below the 25% level there should be evidence for completion by both of the operational definitions discussed previously.  Method Subjects. Six female and four male right handed subjects were drawn from the same subject pool as in previous experiments.  Early Completion 33 Stimuli. There were twenty-eight stimulus configurations. Twentyfive of these corresponded to a five by five factorial design where the two factors were condition (whole, occluded, notched, occluded-t, wholet) and level of occlusion (3%, 5%, 13%, 25%, 46%). The remaining three conditions were the control conditions, single, double, and touch. In the touch condition the target and distractor did not overlap but were in direct contact. All other details were the same as in Experiment 2a. Apparatus. All details were the same as in previous experiments. Procedure. There were several changes from the previous experiment. A fixation point was added for 332 ms before the target presentation and during the target presentation to further reduce eye movements. As well, the number of blocks and trials was changed to 15 blocks of 60 trials. Results After removing the first block and first three trials of each subsequent block there were 798 trials per subject to be analyzed. Each subjects mean reaction time was submitted to a five by five repeated measures ANOVA where the two factors were level of occlusion (3, 5, 13, 25, 46 percent occlusion) and condition (whole, occluded, notched, occluded-t, whole-t). Both main effects and the interaction were significant [F(4,36)=6.6, MS e =704.7, p<.0005 for level of occlusion;  Early Completion 34 F(4,36)=13.4, MS e =946.5, p<.0001 for condition; F(16.144)=4.1, MS e =496.6, p<.0001 for the interaction of the two effects] (See Figure 8 and Table 3). In testing the a priori predictions regarding the small gap sizes it was found that for 3% occlusion the whole, occluded and notched conditions did not differ [t(144)=-1.7 for occluded vs. whole; t(144) =-1.6 for occluded vs. notched; t(144) =0.2 for whole vs. notched]. Thus subjects did not find the three percent notch disruptive in any condition. At five percent the notched condition was slower than the whole condition [t(144)=-2.4, p<.05] but not than the occluded condition [t(144)=-1.3]. The whole and occluded conditions did not differ [t(144)=1.1]. At thirteen percent the occluded and whole conditions did not differ [t(144)=1.5] and both were faster than the notched condition [t(144)=-4.4, p<.0001 for whole; t(144)=-2.9, p<.005 for occluded]. This is the first indication that completion is occurring. At twenty-five percent the notched condition was slower than the occluded condition [t(144)=-2.6, p<.05] and the whole condition [t(144)=-5.0, p<.0001]. The occluded condition was slower than the whole condition [t(144)=2.4, p<.05]. Once again, we have some evidence for completion. At 46% occlusion the notched condition was slower than both the occluded condition and the whole condition [t(144).8.6, p<.0001 for whole vs. notched; t(144)=6.2, p<.0001 for occluded vs.  Early Completion 35  680 A  660 -  ■  1,7 E  0 E i: c o  .«. 0  640 -  •  6H:t— occluded 0 whole .•—' notched —  620 -  0 e re  A 600 -  580  •  •  A  / ••  •  0 I  .  I  .^I^I  0^10^20^30^40^50 Percent Occluded  Figure 8. Results of Experiment 2b  —  Early Completion 36 notched]. The occluded condition was slower than the whole condition [t(144)=-2.3, p<.05]. Note, however, that the difference between the occluded and whole conditions for the 46% occlusion (62ms) is much greater than at 25% occlusion (24ms). This difference between the occluded condition at 46% and 25% occlusion is significant [t(144)=-3.3, p<.01]. The control conditions occluded-t and whole-t did not differ from their respective experimental conditions for any level of occlusion [t(144)<1.0 for all comparisons]. The error analysis revealed no significant main effects and no interaction [E(4,36)=1.4, MS e =.001 for level of occlusion; E(4,36)=0.8, MS e =.001for condition; F(16.144)=0.8, MS e =.001 for the interaction of the two effects]. Discussion  It was predicted that highly occluded objects would not be completed quickly while partly occluded objects would be. In Experiment 2a it was shown that highly occluded objects are not completed while marginally occluded objects (25%) were partly completed. In Experiment 2b it was shown that partly occluded objects are completed and marginally occluded objects are again partly completed. There seems to be an  Early Completion 37 important break point, for these stimuli, at the 25% occlusion level such that below this level there is clear evidence for the operation of early completion while above this level there is clear evidence for failure of completion. At 25% the data is equivocal. By the first operational definition for completion there is a failure while by the second operational definition completion succeeds. It is thus concluded that some information is passed forward to later vision processes in this intermediate condition. It is interesting to note that overall reaction times in Experiment 2b are longer than those in Experiment 2a (Compare Table 2 and Table 3). This 110ms effect did not seem to change the pattern of results and had minimal effects on the difference between the single and double control conditions (29ms in 2a and 26ms in 2b). This effect can be attributed to the addition of the fixation point before and during the presentation of the stimuli. This central fixation point may have had two related effects. First, overall number of eye movements may have been reduced by giving subjects something on which to fixate. Second, some attention may have been diverted from the task to maintain this fixation which would have an effect on the identification process but should not affect the preattentive completion process. This may be taken as suggestive evidence  Early Completion 38 that manipulations of later attentional processes do not affect completion as was proposed in the introduction. Having shown that the early completion processes is spatially limited and further having delineated the extent of this limitation, the next experiments were conducted to show that this result is not task dependent.  Experiment 3 It has been proposed that partly occluded objects are completed early in the visual stream and further that this completion process is spatially (Experiment 1 and 2) and temporally (Sekular & Palmer, 1992) limited. The next two experiments explore how separable this mechanism is from other visual routines. The current experiment investigates effects of changing the subjects' task. If completion occurs early in the visual stream as has been proposed, then we should expect the same spatial limitations regardless of task. As well, in the two previous tasks (matching and identification) it was to the subjects' advantage to complete the object. That is to say, reaction times would decrease to the extent that subjects completed the target. The current task, deciding whether an object was retinally complete or incomplete, requires that  Early Completion 39 the subject not complete partly occluded objects. To the extent that the completion processes is automatic subjects' reaction times in the occluded condition should increase when completion occurs. This test of automaticity is similar to that used by Enns and Rensink (in press) in the search experiment discussed earlier. Thus, there are two purposes for the current experiment: to determine whether the spatial limitations proposed for completion are task independent of other response relevant processes, and to further investigate the automatic nature of completion. The operational definitions of completion must be redefined for this new task. There are again two definitions. First, completion will be assumed when the occluded condition is slower than the notched condition. In these two conditions the retinal stimulation of the two fragments is identical as is the required response. The only difference between these conditions is the extent to which completion can be accomplished (i.e. completion occurs in the occluded condition and not in the notched condition). The second definition states that completion will be assumed when the occluded condition is slower than the whole condition. For these two conditions, when completion occurs, the internal representation is similar, but the response required is different  Early Completion 40 It was predicted that reaction times for the notched and whole conditions would not differ and should be relatively unaffected by the level of occlusion. To reiterate, the occluded condition should take significantly longer than the whole and notched conditions for small gap sizes but not for large gap sizes. In addition, the point at which completion begins to fail should occur at approximately the same level of occlusion as in the previous experiment (i.e. 25% occlusion). Method Subjects. Four right handed males and six right handed females  participated for course credit. Subjects were drawn from the same subject pool as for previous experiments. Stimuli. The item configurations were identical to Experiment 2a  with the exception that the levels of occlusion sampled included 5, 13, 25, 36, 54 percent occlusion. As a reminder, the conditions included were whole, occluded, notched, occluded-t, whole-t, single, and double. Apparatus. The equipment and software used were the same as  previous experiments. Procedure. Overall procedures, display durations, and feedback were  identical to Experiment 2b.  Early Completion 41 The task was to indicate whether the black item was complete or incomplete. Subjects used the index finger of one hand on the keyboard to indicate complete (whole) and the index finger of the other hand to indicate incomplete (notched). The side of responding was counterbalanced across subjects. It was emphasised that the items in the occluded condition were retinally incomplete and thus required the same response as the notched conditions. There were 15 blocks of 60 trials each.  Results After the first block of trials and the first three blocks of each subsequent block were removed there were 798 trials for each subject. The mean reaction times for all correct responses were submitted to five by five repeated measures ANOVA where the two factors were level of occlusion (5, 13, 25, 36, 54 percent occlusion) and condition (whole, occluded, notched, occluded-t, whole-t). Both main effects were significant as was the interaction of the two [E(4,36)=36.6, MS e =875.0, p<.0001 for level of occlusion; F(4,36)=6.4, MS e =4377.8, p<.0005 for condition; E(16,144)=506, MS e =847.8, p<.0001 for the interaction of the two effects].  Early Completion 42  Figure 9. Results of experiment 3  Early Completion 43 In testing the a priori predictions it was found that the occluded condition was significantly slower than the notched condition for all levels of occlusion except the largest [t(144) =6.5 for 5%; t(144)=4.1 for 13%, t(144)=3.7 for 25%; t(144) =2.7 for 36%, p<.0005 for all comparisons; t(144)=1.53 for 54%]. This indicates that the occluded object was completed for all levels of occlusion except for the 54%. The occluded condition was also significantly slower than the whole condition for the first three levels of occlusion [t(144)=8.2 for 5%; t(144)=3.4 for 13%; t(144)=3.6 for 25%, p <.0005 for all comparisons; t(144)= 1.1 for 36%; t(144)=0.3 for 54%]. The whole and notched conditions did not differ from each other for any level of occlusion [t(144)<1.9 for all levels] (See figure 9 and Table 4). The control conditions occluded-t and whole-t did not differ from their respective experimental conditions except at the 25% occlusion level where the occluded-t was faster than the occluded condition [t(144)=2.2, p<.05]. As well, at this level the occluded-t condition did not differ from the notched [t(144)=1.5], whole [t(144)=1.3], or whole-t [(144)=1.0] conditions indicating that for this control condition no completion was occurring.  Early Completion 44 The error analysis indicated two significant main effects and a significant interaction [F(4,36)=9.7, MS e =.001, p<.0001 for level of occlusion; E(4,36)=6.0, MS e =.002, p<.0009 for condition; .E(16,144)=4.7, MS e =.001, p<.0001 for the interaction of the two effects]. The interaction was due solely to the high error rate in the occluded and occluded-t conditions for the 5% occlusion level. Discussion  As was predicted, completion occurred for small gap sizes (<36%) but not for large gap sizes (54%). This shows that the spatial limitations proposed for the completion mechanism are independent of task demands. Although there was evidence for completion at higher levels of occlusion, there are signs of difficulty for completion at the 25% level when looking at the occluded-t condition. The similarity of results between Experiment 2 and 3 support the notion that completion is a separate process from that used to identify or catagorize these target items. Separable does not imply full independence. At present we do not have a precise enough definition of the involved processes to investigate the independence or interaction of these processes. The separable nature of the proposed object completion is still tentative. To strengthen the argument for separability the fourth  Early Completion 45 experiment was run using both static and moving objects to test if motion would affect completion while leaving other processes unaltered. Experiment 4 There are several experiments in the literature that have investigated the interaction of completion and movement. Kellman & Shipley (1992) have drawn a tentative link between motion induced subjective contours and static subjective contours and have further implied a single mechanism operating in these two situations. As well, Gibson and collogues have shown that depth segregation can be accomplished within displays where the only cue to an occlusion contour is motion of random dots (Gibson, Kaplan, Reynolds, & Wheeler, 1969; Kaplan, 1969). More recently Shimojo, Silverman & Nakayama (1989) have shown that completion occurs more readily in moving displays than in static displays. The added information from the accretion and deletion of surface properties at the occlusion boundary must aid in the segregation of the two objects into separate depth planes which should aid in the task of completion. I will be using motion as a tool to show that early completion is separable from later categorization and response based processes. It is  Early Completion 46 predicted that completion will be more effective on moving targets due to the added information provided at the occlusion boundary. At the same time there should be no difference between the moving and static targets in the whole condition. Differences may surface in the notched condition due to the production of illusory contours induced by the motion of the target behind an invisible occluder. The important comparisons to be made for the purposes outlined above are between the static and moving target for the occluded condition. As well, a set of conditions was added where all of the elements in the display move in concert with the target. This display has all the same information as the moving target conditions without the accretion and deletion at the occlusion boundary. For this reason the conditions in which all of the items move in unison may serve as better controls for the target move conditions than the static conditions. Method Subjects. Fourteen subjects were selected from the same subject  pool used previously. Two subjects were discarded: the first failed to complete all of the blocks, and the second had error rates above 25% in the second session. This left nine right handed females and three right handed males. All subjects had normal or corrected to normal vision.  Early Completion 47  Figure 10. Sample Stimuli used in Experiment 4  Early Completion 48 Stimuli.^The item configurations were identical to Experiment 3  with exception that the levels of occlusion sampled included 5, 20, 28, 39, 54 percent occlusion. The conditions included were whole, occluded, notched, occluded-t, whole-t, and double. There was an added factor of motion which had three levels including static, all move, and target move. The static displays were a replication of the previous experiment. In the all move conditions every visible contour of the target and distracting item moved in concert (See figure 10). Thus there was no accreation and deletion of information at the occludion boundary. In the target move conditions only the target moved and all other objects, visible or not were stable. Thus, there was accreation and deletion of information at the occlusion boundary even in the notched condition where where there was no visible occluder. This condition in particular was perceptually odd because of the compelling percept of a complete item moving behind a "phantom" occluder. The motion was in a direction perpendicular to the line on which the distractor could be presented (remember that the distractor could be presented either to the right and above or left and below a right target and to the left and above or right and below of a left target). Thus a right target moved along the diagonal 45° to the right. A left target moved along the diagonal 45° to the left.  Early Completion 49 The items were presented, sequentially, in three positions on each trial to give the impression of motion. The percent occlusion remained constant through all three of the target positions. The target oscillated at a rate of 2.8° per second and moved over a total distance of 0.2°. The stimuli were presented for a total of 265 ms. The added 65ms in relation to previous experiments was necessary due to machine limitations and the desire to have all three positions represented equally. Apparatus. All equipment and software was the same as in previous experiments. Procedure.^Overall procedure was the same as in previous experiments with the exception that subjects ran in two session each consisting of 20 blocks of 50 trials. The added session and blocks were necessary due to the added conditions. Results The first block of both sessions and the first three trials of all other blocks were discarded as practice. This left 1786 trials per subject to be analyzed. The error free reaction times were submitted to a three factor repeated measures ANOVA where the factors were motion (static, all move, target move), condition (whole, occluded, notched, occluded-t, whole-t), and percent occlusion (5%, 19%, 28%, 39%, 54%) . The main  Early Completion 50 effects of condition and occlusion were significant while that for motion was not [F(4,44)=5.2, MS e =9313.8, p<.002 for condition; F(4,44)=47.1, MS e =1317.1, p<.0001 for occlusion; E(2,22)=0.1, MS e =730.6 for motion]. These effects were moderated by two higher order interactions. The first between motion and condition [F(8,88)=2.4, MS e =909.4, p<.05] and the second between condition and occlusion [E(16,176)=7.3, MS e =924.2, p<.0001]. The three way interaction between motion, condition and occlusion was not significant [.E(32,352)=1.2, MS e =824.8]. The motion by condition interaction was due primarily to differences between the target move and the all move conditions for the occluded trials and the whole-t trials. For the occluded conditions, target move was slower than all move [t(96)=-2.2, p<.05] while for the whole-t conditions, target move was faster than the all move [t(96)=2.6, p<.05]. The differences in the whole condition and the occluded-t conditions were in the same direction but not significantly different. Thus collapsing across levels of occlusion subjects found the occluded conditions more difficult when the target only moved as opposed to when all the items moved. The static conditions were intermediate to the other two conditions and not different from either. This thus confirms our a priori  Early Completion 51 predictions that the occluded object would be best completed in the target move conditions. When considering the occluded condition in more detail by examining across the five levels of occlusion, it was found that only for the 20% occlusion level was the target move case slower than the other two motion cases R(352)=-2.3, p<.05 for the static; t(352)=-3.3, p<.01 for the all move]. This again confirms the prediction that completion is more efficient in the target move case for this level of occlusion (see Figure 10 and Tables 5, 6 and 7). The errors from each condition were submitted to the same ANOVA. Again the main effects of condition and occlusion were significant [E(4,48)=3.8, MS e =15.9, p<.01 for condition; .E(4,48)=4.5, MS e =12.5, p<.01]  for occlusion].^As well the interaction of condition and occlusion was significant [E(16,176)=2.9 MS e =9.6, p<.001].^In general the errors followed the same trends as the reaction times and were not further analyzed. Practice effects were analyzed by splitting each of the two sessions in half and submitting the resulting four blocks of data to a four factor repeated measures ANOVA where the four factors were practice, motion, condition, and occlusion. There was an overall effect of practice  Early Completion 52 [F(3,33)=6.2, MS e =71828.0, p<.002].^As well, practice interacted with condition [E(12,132)=4.4, MS e =7944.0, p<.0001] and with occlusion [F(12,132)=12.7, MS e =5308.7, p<.0001]. As well, there was a three way interaction between practice, condition and occlusion [.E(48,528)=2.6, MS e =4070.7, p<.0001]. These higher order interactions were due to an unexpected reversal for the 28% and 39% occlusion such that early in both sessions subjects found the notched and whole conditions difficult and the occluded condition easy. In the latter half of both session the notched and whole conditions became much easier while the occluded condition remained at about the same level. This uncharacteristic reversal was not present in any of the past experiments and is thus attributed to the introduction of motion. At present the only explanation that can be conjectured is one in which subjects are relying on later processes for these boundary conditions where completion is beginning to fail and that these later cognitive processes do not follow the rules proposed for the early completion process. Note that by 54% occlusion the results are fairly stable and in the predicted direction so that the difficulty in interpretation lies only at the 28% and 39% occlusion where subjects' may be using a mix of strategies, both within a given trial and across trials, based on the amount of information available from early processes.  Early Completion 53  Occluded Condition For All Motion Cases 600 -  Static  —0 580 Iii E  —  All in Motion  - I=1- Target in motion  Whole conditions  560 -  d  I.-  •  O  540 -  ^  co e cc^520 -  500 -  480  I^1^1^1^1 0^10^20^30^40^50^60  Percent Occluded  Figure 11. Results from experiment 4  Early Completion 54 Discussion  It had been predicted that the introduction of motion for the target only would enhance the completion process and thus make occluded trial more difficult for these trials when compared to the all move conditions and to the static conditions. This was confirmed. The effect was primarily limited to the 20% occlusion level. This further supports the proposal that early completion is separable from later decision processes. This separability should not be considered a strong form of independence as there was some interaction between early and late processes across practice. The exact nature of the practice effects await further experimentation.  Early Completion 55 General Discussion The question posed at the beginning of this study dealt with whether or not all objects are completed and what spatial factors influence the failure of completion. Sekular and Palmer (1992) showed that completion failed when there was insufficient time allowed for foveated figures. Experiment 1 showed that completion fails for non-foveated targets presented for 200ms with large gap sizes as opposed to small gap sizes which were completed. Experiment 2 expanded on these results to show that at about 25% occlusion there is only partial completion of the occluded object. Beyond this level there was a complete failure of completion. Experiment 3 replicated this failure of completion using a different task (categorization as complete or incomplete as opposed to an identification task). The task irrelevant nature of the completion failure is taken as support for the separability of early completion from later response dependent processes. Experiment 4 demonstrated that early completion is enhanced by placing the target in motion relative to the background items while leaving later discrimination processes, for the most part, unaffected. Overall, completion can be characterized as a rapid, automatic, and spatially parallel process that begins to break down when large gap sizes  Early Completion 56 are presented for short durations. In the real world this implies that early vision attempts to complete partly occluded objects and is, for the most part, successful. Difficulties in completion arise when the object to be completed is highly occluded or only briefly visible. In these cases where there is a failure of completion, the observer must either delay responding or direct attention to aid in the interpolation of the retinal fragments. The exact role that attention plays in completion was not addressed in this paper. However, a comparison of Experiments 2a and 2b seems to indicates that completion is independent of attentional demands. This begs further experimentation to ascertain the benefits that attention could lend to completion. Current theories of visual information processing that utilize the early-late dichotomy have typically thought of the first stage as registering only simple 2-D features as they appear on the retina. The latter stages are thought to integrate these free floating features by directing attention to a given location in space. (Treisman, 1988; Treisman & Gelade, 1980; Treisman & Sato, 1990). According to Feature Integration Theory (Treisman & Gelade, 1980), the interobject relation of occlusion should not be computed early in vision since this attribute requires the computation of spatial relations between objects which are  Early Completion 57 themselves not registered until later processing stages. Thus, the demonstration that occluded objects are completed preattentively (Enns & Rensink, in press) counters the simplistic view of early vision. Instead, early vision should be conceptualized as a hard wired system that computes scene-based 3D attributes that in the past have been shown to be relevant (Enns, 1992). I am argueing here not for a simple addition of new 3D features to the already present framework of 2D features (Enns, 1990) but for a reconceptualization of what are the primitive elements of perception (Enns & Rensink, 1993). Support for this line of argument comes not only from studies of occlusion (Enns, 1992; Enns & Rensink, in press), but also from search studies showing preattentive registration of shape from shading, direction of lighting and the recovery of slant from texture (Aks, 1993; Aks & Enns, 1992; Enns, 1990; Enns & Rensink, 1992; Enns & Rensink, 1993; Kleffner & Ramachandran, 1992). These 3D attributes of the scene can not be derived from the mere registration of 2D features but instead require the computation or recognition of certain complex invariant attributes in the environment. The fact that early vision is sensitive to these attributes speaks against the simplistic view presented in Feature Integration Theory. The current work follows from the finding that occlusion relations  Early Completion 58 are registered preattentively. For an attribute of the image or the scene to be computed in parallel across the visual scene, there must be some operator that acts at every location of the image over a finite spatial area, called a neighbourhood of operation (Enns, 1992; Enns & Rensink, 1993; Zucker, 1987). The assumption that completion is spatially limited was tested and found to be true, adding further support to the notion that occlusion relations are registered early within finite neighbourhoods of operation. Preattentive processes have for the most part been characterized as automatic data-driven processes that are beyond the subjects voluntary control. Experiments 3 and 4 speak to this issue and support the conjecture that completion is indeed automatic. Recall that in those experiments it was to the subjects advantage not to complete the object. Despite this, subjects completed the occluded figure, thus slowing performance on the task for small gap sizes. Note, however, that the automatic nature of completion does not prevent it being manipulated by scene attributes (i.e. the kinetic nature of the objects in Experiment 4). This last experiment supports the notion that completion is an automatic process that is distinct in terms of operating stages from later attentionbased processing stages. These later stages are presumably used to  Early Completion 59 identify or describe the objects passed down the stream by the early vision processes. As this is still a new area of study, there remain many questions which need to be answered in an objective, performance-based way. The exact relation between alignments of attention, both voluntary and involuntary, and completion need to be further delineated. Specifically, when attention is aligned with an area of occlusion, will completion be improved? To address this question one could use a Posner (1980) cueing paradigm in conjunction with the present notched task to discern if on valid trials completion is more efficient than on neutral or invalid trials. An additional area of study that requires further research is the exact time course of the completion process. Sekular and Palmer (1992) showed that partly occluded objects are completed by 200ms but they used an indirect priming paradigm and required that a third object be present in the probe to eliminate apparent motion affects. As well, they only used one level of occlusion. Are the more highly occluded objects not completed at all, as implied earlier, or do they simply take more time to complete? This question could be addressed by conducting a microgenesis experiment similar to the earlier work (Sekular & Palmer, 1992) and using primes of different levels of occlusion.  Early Completion 60 The finding that the early completion process can be manipulated independently of the later, presumably attentional, process begs the question: Can the reverse be done? Can the attentional process be manipulated while leaving early completion unaffected? To address this question, one could manipulate the complexity of items to be identified or matched while leaving the amount of occlusion unaltered. It is predicted that the effect of complexity would be orthogonal to the effect of amount of overlap. This would further speak to the question of the role that attention plays in completion. The evidence presented in the current paper is consistant with a local completion process. That is, one in which completion is based on the attributes present at the occlusion boundary (Kellman & Shipely, 1992). An alternate view, where completion is affected by global attributes of the figure to be completed, has been recently presented by A. Sekular (submitted). According to this theory, there are two factors contributing to the completion of occluded objects: a top-down influence based on object symmetry and a bottom-up input based on local junction information. This hybrid model has a certain intuitive appeal but at present needs further refinement. Evidence for the global aspect of completion comes from high speed priming studies where the degree of  Early Completion 61 symmetry and the axis of symmetry were manipulated. It was found that for fourfold symmetrical objects completion was entirely predicted by global theories (Sekular & Gaber, 1993). When the symmetry was only twofold or onefold, the axis of symmetry determined the mix of global and local contributions with local global theories dominating when the symmetry was across the vertical or oblique axis but not the horizontal axis. Due to the indirect nature of priming paradigms it would be appropriate to conduct further research into the exact contribution of global and local factors into the completion process. An additional note for this area of research is the exact connotations from the terms global and local which are in dire need of stringent definition. One final area which requires investigation is the role that depth segregation or figure-ground relations plays in completion. To individuate and recognize an object, two things must occur: the figure must be separated from the ground and the figure must then be completed or interpolated. The ordering of these two processes is a question that must be answered. The current study makes no mention of depth or segregation in depth. This is because the work presented does not address this question, although it is intimately related. In closing, it seems that there is still a great deal of work to be done  Early Completion 62 before we can understand completion and its role in everyday perception. There is strong support now that completion does occur early in visual processing and in parallel across the scene. Further attributes of this process and the delineation of other potential completion processes awaits further work.  Early Completion 63 References Aks, D.J. (1993). The Analysis of Slant-From-Texture in Early Vision. Unpublished doctoral dissertation, University of British Columbia, British Columbia, Canada. Aks, D.J., & Enns, J.T. (1992). Visual Search for Direction of Shading Is Influenced by Apparent Depth. Perception & Psychophysics, 52, 6374. Enns, J.T. (1990). Three-dimensional features that pop out in visual search. In D. Brogan (Ed.) Visual Search. Taylor & Francis: London, pp. 37-45. Enns, J.T. (1992) The Nature of Selectivity in Early Human Vision. In B. Burns (Ed.), Percepts. concepts. and categories: The representation and processing of information (pp. 2-34). Amsterdam: Elsevier Science Publications. Enns, J.T., & Rensink, R.A. (1993). A model for the rapid interpretation of line drawings in early vision. In D. Brogan, A. Gale, K. Carr (Eds.) Visual Search 2. London: Taylor & Francis, pp.73-89. Enns, J.T., & Rensink, R.A. (in press). An Object Completion Process in Early Vision, To appear in A.G. Gale (Ed.), Visual Search III: Proceedings of the Third International Conference on Visual Search. London: Taylor & Francis. Gerbino, W., & Salmaso, D. (1987). The Effect of Amodal Completion on Visual Matching, Acta Psychologica, 65, pp. 25-46. Gibson, J.J., Kaplan, G.A., Reynolds, H.N., Jr., & Wheeler, K. (1969). The change from visible to invisible: A study of optical transitions. Perception & Psychophysics, 5, 113-116. Greenhouse, S.W., & Geisser, S. (1959). On methods in the analysis of profile data, Psychometrika, 24, 95-112.  Early Completion 64 Huynh, H., & Feldy, L., S. (1973). Conditions under which mean square ratios in repeated measurement designs have exact F-distributions. Journal of american Statistical Association, 65, 1582-1589. Kanizsa, G., & Gerbino, W. (1982). Amodal Completion: Seeing or Thinking?. In J. Beck (Ed.) Organization and Representation in Perception. Hillsdale, NJ: Erlbaum. Kaplan, G.A. (1969). Kinetic disruption of optical texture: The perception of depth at an edge, Perception & Psychophysics, 6, 193-198. Kellman, P.J., & Shipley, T.F. (1992). A Theory of Visual Interpolation in Object Perception, Cognitive Psychology, 23, 141-221. Kleffner, D.A., & Ramachandran, V.S. (1992). On the Perception of Shape from Shading, Perception &Psychophysics, 52, 18-36. Posner, M.I. (1980). Orienting of Attention, Quarterly Journal of Experimental Psychology, 32, 3-25 Ramachandran, V.S. (1992). Filling in Gaps in Perception: Part I, Current Directions in Psychological Science, 1, 199-205. Ramachandran, V.S. (1993). Filling in Gaps in Perception: Part II. Scotomas and Phantom Limbs, Current Directions in Psychological Science, 2, 56 Rensink, R.A., & Enns, J.T. (1992). VScope Software. micropsych software, U.B.C. Sekular, A. (submitted). Perceptual Completion of Partly occluded Objects: Evidence from the primed matching paradigm. Perception. Sekular, A.B., & Gaber, E. (July, 1993). Perceptual completion: A Local or Global Process? Paper presented at the Third Annual Meeting of the Canadian Society for Brain, Behaviour, and cognitive Science: Toronto, Ontario.  Early Completion 65 Sekular, A.B., & Palmer, S.E. (1992). Perception of Partly Occluded Objects: A Microgenetic Analysis, Journal of Experimental Psychology: General, 121, 95-111. Shimojo, S., Silverman, G.H., & Nakayama, K. (1989). Occlusion and the Solution to the Aperture Problem for Motion. Vision Research, 29, 619-626. Treisman, A. (1988). Features and Objects: The Fourteenth Bartlett Memorial Lecture, The Quarterly Journal of Experimental Psychology, 40A, 201-237. Treisman, A., & Gelade, G. (1980). A feature Integration Theory of Attention. Cognitive Psychology, 12, 97-136. Treisman, A., & Sato, S. (1990). conjunction search Revisited, Journal of Experimental Psychology: Human Perception and Performance, 16, 459-478. Zucker, S.W. (1987). Early Vision. In S.C. Shapiro (Ed.), The encyclopedia of artificial intelligence (pp. 1131-1152). NY: John Wiley.  Early Completion 66 Appendix A - Raw Tabled Data Table 1. Results of Experiment 1: Raw Data (RT and Error) ^ 67 Table 2. Results of Experiment 2a: Raw Data (RT and Error) ^ 68 Table 3. Results of Experiment 2b: Raw Data (RT and Error) ^ 69 Table 4. Results of Experiment 3: Raw Data (RT and Error) ^ 70 Table 5. Results of Experiment 4 - Static Conditions: Raw Data (RT and Error) ^  71  Table 6. Results of Experiment 4 - All Move Conditions: Raw Data (RT and Error) ^  72  Table 7. Results of Experiment 4 - Target Only Move Conditions: Raw Data (RT and Error) ^  73  Early completion 67  Table 1. Results of Experiment1: Raw Data Reaction Times Occlusion Conditions Whole Occluded Occluded-truncated Notched Different  Gap Size Large  Small  722 853 713 1006 971  (42) (38) (52) (58) (40)  733 748 744 899 938  (39) (48) (41) (69) (45)  Type of Condition Control^Conditions  Single  Double  Same Different  615 (27) 721 (32)  751 (51) 949 (57)  Error Rates Gap Size Occlusion Conditions  Small  Large  Whole Occluded Occluded-truncated Notched Different  6.2 (2.0) 10.0 (4.5) 6.9 (2.9) 23.2 (6.1) 7.2 (1.4)  6.9 (3.1) 15.3 (2.7) 5.7 (1.6) 42.3 (8.8) 9.0 (1.5)  Type of Condition Control Conditions^Single^Double Same Different  1.3 (0.7) .^5.3 (2.4)  7.5 (2.0) 8.4 (3.0)  Early completion 68  Table 2. Results of Experiment 2a: Raw Data Percent Occluded Conditions^25%^50%^70%^85%  Reaction Time Whole Occluded Notched Notchedalone Whole-T  470 (18) 496^(19) 514^(19) 486 (21) 465 (19)  471 515 515 506 484  (17) (19) (20) (20) (21)  475 540 534 512 463  (18) (19) (20) (17) (20)  486 (20) 570 (23) 573 (24) 535 (28) 486 (24)  Error Rates Whole Occluded Notched Notch-alone Whole-T  4.2^(1.4) 2.7 (0.9) 1.9^(0.9) 1.9^(0.8) 2.5 (0.8)  1.1 3.6 2.3 2.5 2.0  ^ (0.6) 1.4 (0.5) ^ (1.0)^4.0 (1.3) (1.2) 3.6 (1.6) ^ (0.8) 2.8 (0.6) ^ (1.0) 0.6 (0.4)  The control condition had the following RTs and Error rates: Single: 455 (18), 1.5 (0.4) Double:484 (19), 2.0 (0.6)  2.2 (1.7) 8.2 (1.7) 4.2 (1.1) 3.4 (0.8) 2.5 (1.1)  ^  Early completion 69  Table 3. Results of Experiment 2b: Raw Data Percent Occluded 5%  3%  Conditions  13%^25%^46%  Reaction Time (23) ^(20)^581 586 590 (20) 607 (25) 588 (20) Whole 610 (20)^643 (25) 605 (22) 590 (21) 598 (23) Occluded 635 (23)^667 (17) 633 (23) 605 (21) 612 (21) Notched 601 (21)^624 (19) 602 (23) Occluded-T 590 (21) 599 (22) (21) ^(24)^592 593 590 (20) 603 (23) 589 (21) Whole-T  Error Rate Whole Occluded Notched Occluded-T Whole-T  2.5 3.8 2.1 0.8 1.9  (0.9) (1.7) (0.6) (0.5) (1.3)  2.4 2.8 3.7 3.2 3.1  (0.7) (1.3) (1.3) (1.1) (1.0)  3.8 2.8 2.2 3.2 3.3  (1.0) (1.5) (0.8) (1.0) (1.1)  1.7 2.7 3.9 3.2 3.5  (0.8)^3.1 (1.2) (1.1)^3.8 (1.4) (1.6)^4.7 (1.4) (1.1)^4.7 (1.1) (1.6) 2.9 (1.3)  Note the control condition rt and error rates were as follows 567 (18), 2.2 (0.9) Single: 593 (24), 0.6 (0.4) Double: 588 (22), 4.6 (1.1) Touch:  Early Completion 70  Table 4. Results of Experiment 3: Raw Data Percent Occluded Conditions^5%  13%  25%^36%  54%  Reaction Time Whole^675 Occluded^782 Notched^698 Occluded-T 787 Whole-T^686  (47) 683 (39) (38) 727 (41) (43) 674 (41) (38) 742 (39) (40) 671 (67)  668 713 665 685 672  (41) (42) (35) (37) (38)  666 680 645 673 669  (37) (40) (43) (36) (34)  670 (35) 674 (40) 654 (38) 669 (38) 661 (37)  Error Rate Whole^0.7 (0.4) Occluded^7.8 (2.7) Notched^1.6 (0.7) Occluded-T 12.7 (3.8) Whole-T^1.8 (0.6)  1.3 2.0 1.1 4.6 2.4  (0.7) (1.2) (0.6) (2.1) (1.1)  1.6 2.0 0.4 2.9 0.4  (0.7) (0.9) (0.4) (0.9) (0.4)  1.0 1.0 0.7 1.8 2.0  (0.5) (0.5) (0.5) (0.8) (1.3)  0.6 (0.4) 0.7 (0.5) 0.8 (0.5) 0.0 (0.0) 0.7 (0.7)  Note the control condition rt and error rates were as follows Single: 567 (18), 2.2 (0.9) Double: 593 (24), 0.6 (0.4) Touch: 588 (22), 4.6 (1.1)  ^  Early Completion 71  Table 5. Results of Experiment 4 - Static Conditions Percent Occluded Condition  5%  20%  ^  28%  ^  39%  ^  54%  Reaction Time whole 541 (26) (30) ^521 (18)^534 occluded 552 (32)^531 (28) 598 (36) notched 527 (20) 513 (26)^496 (28) occluded-t 579 (25) 551 (29)^534 (36) whole-t 549 (22) 523 (16)^532 (22)  527 (22) 528 (25) 484 (27) 508 (25) 520 (17)  517 515 492 507 532  (15) (22) (27) (26) (22)  2.9 1.3 0.0 0.6 0.9  0.6 2.1 1.3 1.2 1.9  (0.4) (1.0) (0.9) (0.6) (0.7)  Error Rate 1.6 whole occluded 4.1 notched 1.3 occluded-t 3.7 whole-t^2.4  (0.8) (1.7) (0.7) (1.6) (0.8)  2.3 1.2 2.8 1.6 2.5  (1.0) (0.7) (1.3) (0.9) (1.4)  1.9 2.5 1.0 2.2 3.9  (0.7) (1.8) (0.7) (1.0) (1.3)  (0.9) (0.7) (0.0) (0.6) (0.5)  ^  Early Completion 72  Table 6. Results of Experiment 4 - All Move Conditions Percent Occluded 5%^20%^28%^39%^54%  Reaction Time whole 529 (16) ^(22)^520 534 (20) 595 (27) occluded 539 (37)^537 (28) notched 531 (27) 517 (30)^502 (30) occluded-t 596 (27) 550 (30)^517 (23) whole-t 544 (18) 540 (12)^538 (20)  514 (12) 513 (25) 497 (30) 507 (22) 539 (22)  516 524 484 506 522  (21) (25) (22) (24) (18)  2.1 1.9 1.7 1.7 1.4  2.7 1.2 1.2 1.6 0.6  (0.9) (0.7) (0.9) (0.9) (0.4)  Error Rate whole 2.2 occluded 6.8 notched 0.4 occluded-t 2.3 whole-t^2.8  (0.9) (1.7) (0.4) (1.9) (1.0)  1.5 2.1 2.1 2.1 1.8  (0.8) (0.8) (1.0) (1.3) (0.6)  1.9 2.2 1.2 2.8 3.1  (0.7) (1.0) (0.6) (1.2) (1.0)  (1.0) (1.4) (0.7) (1.0) (0.7)  ^ ^  Early Completion 73  Table 7. Results of Experiemnt 4 - Target Only Move Conditions Percent Occluded 5%  28%  20%  ^  39%  54%  Reaction Time 526 whole occluded 599 525 notched occluded-t 588 whole-t 520  (20) (27) (26) (24) (15)  508 (18)^526 (18) 578 (41)^532 (23) 498 (27)^514 (24) 570(27)^528 (23) 540 (23)^526 (21)  506 (14) 524 (26) 488 (26) 512 (28) 509 (17)  519 530 488 524 515  (20) (26) (24) (33) (17)  2.8 1.7 0.4 0.4 2.1  2.3 2.1 0.8 1.2 1.2  (1.2) (1.0) (0.6) (0.6) (0.6)  Error Rate whole 1.9 occluded 6.0 notched 1.0 occluded-t 4.5 whole-t^0.6  (0.6) (2.0) (0.7) (1.1) (0.4)  1.5 2.7 0.9 5.4 2.4  (0.7) (1.1) (0.7) (1.9) (1.0)  2.5 3.4 1.1 2.3 0.8  (1.3) (1.1) (1.1) (1.2) (0.5)  (0.7) (1.0) (0.4) (0.4) (0.8)  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items