Human Information Systems - Change Detection - Essay Example

HUMAN INFORMATION SYSTEMS CHANGE DETECTION Prof Introduction Change detection primarily involves the visual process in which changes occurring in our environment are detected or apprehended by humans. The natural skill to immediately spot these changes is significant to human being since it plays a very important part in our daily activities. Recognizing the changes in your surrounding such as people trying to cross the street, the car behind you, changing red light to green, and many other transformation help us to adjust in a certain way and act accordingly. There many issues involving change-detection mechanism and several research and experiments are underway to determine the real cause of failures. Some of those important issues are presented in this paper along with our own analysis. The Coglab Experiments in Change-Detection The Coglab experiments on Change Detection are aimed to gather substantial evidence that humans do find it difficult to instantly recognize changes in a visual scene. The idea is based on Rensink (2002) study that it will take a while with some degree of concentration for humans to detect sudden changes in an object. People according to the author don’t actually store much details of a sight in their memory and in order to distinguish an object change, a person need to concentrate on the object. The experiment presented two alternating images occasionally separated by gray field. The requirement is to identify the changes between the two versions of the picture as quick as possible to determine the time it takes for a human to respond and recognize the variation in the image precisely. There are a total of 16 trials with half of it showing images without a change and a half with a “flicker” or a mask. The inclusion of the mask is to prove that it can prevent automatic detection of change and to validate the possibility that a person is indeed need to examine every object in the scene to recognize the changed in the object. Change Blindness According to researchers in the field of psychology, the person’s failure to instantly recognize or detect the changes in an object is caused by change blindness. Change blindness is believed to be the cause of many accidents particularly on the road where drivers glancing away from the road and back cannot instantly detect the changes (Rensink, 2002). A recent study also reveals that humans are in a way insensitive in detecting changes in alternating visual displays (Cohen et. al., 2005) The word change according to Rensink (2002) is about alteration, variation, or transformation of a well-defined object in a certain period of time regardless of size and complexity for as long as the object continues to exist over the sequence of its transformation. There are many varieties of changes occurring in an object such as changes in position or motion which is defined as a sequential disparity of an object at a point in space. Another important form of change is dynamic changes occurring in an object progressively until it reaches a complete change. This is similar to seeing or visually detecting the change while in progress. However unlike changes and motion, objects undergoing dynamic changes or transforming itself does not need to be continually present while the change is in progress. Since the experiment is all about change detection in images variation, it is important to discuss the change and difference in an object. While changes are recognized as a transformation in a period of time, difference is known as the absence of complete similarity in the properties of two or more objects. Although the concept of change and difference shares common elements, they are two distinct activities (Rensink, 2002). The ability to recognize or “spot” the difference between images alongside each other is entirely different from detecting the change in a couple of images consecutively presented. For instance, changes are for single object while difference is to two. Changes include temporal transformation and measures similarity in connection with the same object in another period in time while difference possesses no properties of transformation and its similarity depends on comparing objects that may not exist concurrently (Rensink, 2002). These distinctions are very important because they can affect the nature of experiments considered pertinent to the discussion of change detection. Much evidence pertaining to change blindness shows that majority of observers are unable to recognize the changes in their environment and suggested that it can only occur in the “presence of focused attention” (Thornton et. al., 2000). Overall, these findings suggest that our perception of things around us is unstable when it’s outside the range of human’s focal attention. In other words, people cannot distinguish the changes unless they put forth some effort to examine the object. However, this is not the case in the series of experiment conducted by Thornton et. al (2000) on change blindness. Unaware Change Detection The experiment result suggests that it is often hard for the observers to deliberately recognize changes in visual presentations when they are in a situation where local transients are decreased. Although observers are aware and expecting changes to occur, they still are unable to successfully detect all of them. In a different approach, observer’s exhibits an above chance levels when presented with a forced choice task even though unaware of any change. Thornton et. al (2000) explained that such behaviour is seen as perceptions devoid of awareness or unconscious detection of change in the cognitive literature. To determine the origin of such unconscious change detection below the threshold shift of attention, spatial cueing concept were used but found no evidence directly linking the levels of focal attention. On the other hand, Thornton et. al. (2000) is confident that the hidden portion of attention is not a major factor to the overall result of the experiments. They also believed that unaware change detection is beyond the area of concentration and will eventually create a serious challenge to the validity of attention and spatio-temporal rationality which already has become a major part of theories about scene perceptions. It is now evident that attention is not enough to guarantee change detection and it’s not always necessary because people even while consciously detecting the changes in an object sometimes fail. Finally, in relation to these findings, a report claims that some observers do in a way “sense” that something is changed in the scene which about a number of seconds before they can perceptively recognize the change (Thornton, 2000). Influence of Shape in Change Detection In a study conducted by Cohen et. al. (2005) they came with the result confirming the influence of shape or curvatures of shape in human change detection capabilities. The study reveals that people are more responsive to concave sections of shapes than its convex regions. However, this kind of experiment is too complex and may result in further argument and research. A more practical study by McCarley and Vais (2004) investigated the effects of cognitive distraction on change detection particularly in a common traffic scene. The experiment reveals that mobile phone conversation distracts the natural ability of change detection. Conversation Distracts Change Detection Change blindness occur in various visual events such as flickers, ego motion, occlusion, presence of irrelevant transients signals, and it can happen naturally outside the laboratories in the real world setting. There is a strong possibility that change detection failures are the result of human limitations on perception and cognition thus the study of “cognitive distractions” effects is necessary. In a timed experiment simulating a traffic scene, the participant were given mobile phones and were asked to memorized short sentences while driving. The data reveals that with the burden of a distracting task, the participants change detection capabilities were slower even for changes in the scenes that are highly evocative. This means that distraction in some way spoils human perception even on highly interesting changes in the scene (McCarley and Vais, 2004). Overestimation of Change Detection- Meta-Cognitive Errors Errors in change detection are also a major issue in the psychological community. In a recent research conducted by Levin et. al. (2000), participants failed to detect changes between natural and artificial scenes. It is believed that metacognitive errors arising from participant’s overestimation to which changes will be detected. As majority of the papers regarding the issue indicates that people startlingly fail to recognize large changes to scenes. People are unable to catch the manifestation and desertion of elements in natural and artificial scenes. The findings show that misconception of vision and over confidence to detect changes can greatly affect people’s cognitive capacity resulting in metacognitive errors. For instance, eyewitnesses customarily are very confident of their answers but research in eyewitness memory of dramatic events reveals that there is diminutive relation between confidence and accuracy. In a series of experiments, participants repeatedly overestimated their ability to detect changes. In the digit-span test where psychology students were asked to remember a string of numbers suggest that overestimation is not entirely the cause of responses preconception. This shows that participants are disposed to give a negative response and probably more precise in estimating short-term memory than estimating their own change detection ability. Overall, the large majority of participants who thought they can detect changes failed. The probable reason for this according to Levin et. al. (2000) is the biases in reasoning regarding daily events that made them believed that the few changes they detected are indications of successful and error-free change detection. This further confirms the common notion about the direct “feeling of perception and stability” (Levin et. al., 2000) of the real world tends to create an outlook that infringement of stability would be at once evident. In other words, people would have more success than failures if they always think about success in their daily endeavours since this would have a great effect on reasoning and decision-making. However, failures and errors of overestimation may somehow connect to the impossibility of these changes occurring in real situation and may in fact help to understand the composition of natural events deeper. Since the errors are not based on illogical reasoning, it is reasonable to suppose that these are metacognitive thinking where unusual events are on average have to be noticed. In other situation and despite the variations involving real-world and simulated displays, we may achieve an enhanced positive reception of the essential change detection and for the thoughts that manipulates our perception of the natural environment. Visual and Memory Free-Search Change detection is relatively higher when participants confine their thought to the retrieval and comparison of the target object (Hollingworth, 2003). There is sufficient evidence that visual representations of objects in the scene build up in memory as the participant’s attention and eye are focused on the scene providing enough information to make fine visual conclusion to detect change or variations. Hollingworth (2006) once again emphasizes his theory of visual representation in the brain where images of a scene are temporarily stored. For instance, during the eye movement when the processing of the scene is interrupted, sensory persistence decays instantly and focus is transferred to the next scene. On the other hand, directing the attention to an object enables a coherent visual representation consolidating into a more stable form. After the attention is withdrawn from the object, abstracted visual presentations persists and accumulates in memory as the eye moves away from object to object. This is the phenomenon called memory-free search which revolutionizes the belief on awareness, concentration, and memory comparable to coherence theory. Unattended Changes – Visual Onset and Serial Search Process In another interesting but different study conducted by Smilek et. al. (2000) indicates that unattended changes play useful roles in directing focal attention. Humans tends to often exhibit failure to detect changes between two different scenes or objects and such failures or change blindness are occurring when the presentation of these different scene are separated by a short chronological interval, when these images are changed during an eye saccade, and when the object’s view change to another angle of the same scene similar to movies. The experiment was designed to determine if unattended changes are indeed functional in guiding focal attention. The results were assessed using slopes of the search functions for locating large and small changes and the fundamental assumption was, the shallower the slopes are the more efficient the pre-attentive guidance is to focal attention. The findings is consistent in all three experiments and presented a strong conclusion that unattended information is indeed plays an important role in directing attention to the position of change. This result is somewhat related to previous experiment results that changes between two successive displays are guided by two unknown processes. This means when there is no temporal interval between changes the attention is guided by a sudden visual onset to the location of the change while the opposite is guided by a controlled serial search process. Although both processes are undeniably plays a role in directing the location of change, none of them can explain the current experiment results. The conclusion is therefore maybe stated that attention is not only directed by abrupt onsets and controlled serial process but by unattended information about changes or discrepancies among displays. The implications as Smilek et. al. (2000) is worried about is that unattended information are accumulating in memory between displays. Surprisingly, this assumption is already proven by Hollingworth (2006) in previous paragraphs and therefore conclusions that there are memories for changes even when participants claim they are unaware of the changes are correct. Furthermore, encoded scenes in memory facilitating visual search are also true in Hollingworth (2006) report. Open Issues in Change Detection Finally, with more and more research about change blindness, there are still unsolved ambiguities on how these failures occur. What are the mechanisms involved and what exactly is it? Rensink(2002) believed that the issues concerning change detection is still open for analysis and further study. Some other relevant issues include the distinction between identification and detection of change where identification appears to be more difficult according to several studies. There are issues about the difference between implicit and explicit detection of change and the existence of such form of perception since several experiments already revealed the possibility of unaware detection of change. Importantly, issues on the limitations to which change-detection can retrieve visual images because if change blindness can come from the inability to secure relevant information, there is also a strong possibility that it can come from the inability to retrieve relevant representations (Rensink, 2002). Conclusion It is evident that the Coglab experiments is just one of the many experiments involved in the study of change detection failures. Although it only covers the fundamental test, the effectiveness of such experiment is undoubtedly sufficient in confirming the fact that there are indeed variations and factors affecting human’s change detection ability. Issues and other failure detection test that are presented and discussed in this paper are meant to help improve the Coglab experiment procedures and anticipate that it would sooner or later lead to additions of more test and experiments for change-detection. Our research has made us more aware of the psychological origins of change-detection failures particularly those involving meta-cognitive errors, visual and memory free-search. Although there are still open issues in change detection mechanism, further study and experiments will sure help us understand the real nature of human’s change blindness. References: Cohen Elias et. al., 2005, “What Change Detection Tells Us About the Visual Representation of Shape”, Journal of Vision (2005) 5, 313-321,Department of Psychology and Center for Cognitive Science, Rutgers University, New Brunswick, NJ, USA, Available online at http://journalofvision.org/5/4/3/ Hollingworth Andrew, 2006, Visual Memory for Natural Scenes: Evidence from Change Detection and Visual Search”, Visual Cognition, 2006, 14 (4/5/6/7/8), 781_807, University of Iowa, Iowa City, IA, USA, Psychology Press Ltd, Taylor and Francis Group Hollingworth Andrew, 2003, Failures of Retrieval and Comparison Constrain Change Detection in Natural Scenes, Journal of Experimental Psychology: Human Perception and Performance 2003, Vol. 29, No. 2, 388–403, Copyright 2003 by the American Psychological Association, Inc., 0096-1523/03/$12.00 DOI: 10.1037/0096-1523.29.2.388 Levin et. al., 2000, “Change Blindness Blindness: The Metacognitive Error of Overestimating Change-detection Ability”, Visual Cognition 2000 7 (1/2/3), 397–412, Dept. of Psychology, Kent State University, Kent, OH, USA, Psychology Press Ltd. McCarley Jason and Vais Margaret, 2004, “Conversation Disrupts Change Detection in Complex Traffic Scenes”, Beckman Institute, University of Illinois at Urbana-Champaign, 405 N. Mathews, Urbana, IL 61801, HUMAN FACTORS, Vol. 46, No. 3, Fall 2004, pp. 424–436.Copyright © 2004, Human Factors and Ergonomics Society Rensink Ronald, 2002, “Change Detection”, Departments of Psychology and Computer Science, University of British Columbia, Vancouver, British Columbia, V6T 1Z4 Canada, Annual Rev. Psychology 2002. 53:245–77, Copyright 2002 by Annual Reviews Smilek Daniel et. al., 2000, “Does Unattended Information Facilitate Change Detection?”, Journal of Experimental Psychology: Human Perception and Performance 2000, Vol. 26, No. 2, 480-487, Copyright 2000 by the American Psychological Association. Inc., 0096-1523/00/$5.00 DOI: 10.1037//0096-1523.26.2.480 Thornton Ian et. al., 2000, “Change Detection Without Awareness: Do Explicit Reports Underestimate the Representation of Change in the Visual System?” Cambridge Basic Research, Nissan Research and Development, Inc.,Cambridge, MA, USA, Psychology Press Ltd 2000 Read More