Alex Liu is a rising senior majoring in Biology. He was awarded a Fall 2016 Independent Grant which he used to conduct research on scene processing in the brain under Dr. Daniel Dilks.
Research is the process of looking for something that evidence and theory support, but its existence has yet to be seen. Not only is this is the foundation of why we do research but it is also the reason why research is so hard. In my current research, we believe that the brain has more specialized regions for scene processing than what current research has shown us. Instead of there being one stream for scene processing, we propose that there are actually two: one for navigation and one for categorization. The philosophy behind a two-stream system for scene processing comes from the fact that we can categorize without navigating; however, we cannot navigate through a scene without categorizing what is in it. An example would be trying to walk through a cluttered room. If we want to get to the other side of the room without stepping on anything or walking into anything, we must be able to categorize what objects are in the scene and what type of room it is, and then plot a path to avoid those objects. With this in mind we came up with a two stream processing system.
Currently I am running behavioral studies to see if there is an inversion effect when navigating through a scene upside down vs. right side up scenes (figure 1). An inversion effect will show an “impairment” in the navigation system. This impairment should be seen when analyzing the accuracy of the subjects’ responses. We expect that categorizing and navigating in upright scenes will be matched in accuracy, while navigating in inverted scenes will be much worse than categorizing in inverted scenes (figure 2). Categorizing and navigating in upright scenes should be matched because, without restricting natural human processes, we should be able to navigate through a scene just as easily as they are able to categorize the scene. We expected an inversion effect in navigation because humans have never navigated through a scene upside down before and the brain is not “wired” that way, thus making it much harder than navigating upright. An inversion effect will provide the ground bases of evidence that will eventually allow me to move on to fMRI and show that there indeed are specific regions in the brain allocated towards either navigating or categorizing. Combined behavioral and neural evidence of such a “scene inversion effect for navigation,” but not categorization, will provide the strongest evidence to date that scene processing is indeed composed of distinct systems for navigation and categorization.
The challenging part of my project
right now is trying to change small variables so that the accuracy for
navigating and categorizing upright scenes are the same. Currently subjects are
better at navigating the scenes than categorizing upright scenes. I believe
that the problem lies in the memorization of what button corresponds to what—key
mapping. While the key mapping for
navigation is self explanatory (left finger = go left, middle finger = go middle,
right finger = go right), the mapping for categorizing must be memorized. Moving
forward, in an attempt to resolve this problem, I will extend the categorization training
time for key mapping.
I found it surprising and infuriating how long the behavioral part of this experiment has taken me. It is weird to have worked so long in the forefront, creating and building the outline of the experiment, and not seeing the expected outcome. Science classes have taught me that by following a procedure you will get the correct result, but research has reminded me that science is about exploration and learning. Hopefully, I will be able to fix these problems and move on to fMRI and see if my hypothesis holds neurally as well as behaviorally.
Visit the Undergraduate Research Programs website to learn more about applying for Independent Research Grants.
Research is the process of looking for something that evidence and theory support, but its existence has yet to be seen. Not only is this is the foundation of why we do research but it is also the reason why research is so hard. In my current research, we believe that the brain has more specialized regions for scene processing than what current research has shown us. Instead of there being one stream for scene processing, we propose that there are actually two: one for navigation and one for categorization. The philosophy behind a two-stream system for scene processing comes from the fact that we can categorize without navigating; however, we cannot navigate through a scene without categorizing what is in it. An example would be trying to walk through a cluttered room. If we want to get to the other side of the room without stepping on anything or walking into anything, we must be able to categorize what objects are in the scene and what type of room it is, and then plot a path to avoid those objects. With this in mind we came up with a two stream processing system.
Currently I am running behavioral studies to see if there is an inversion effect when navigating through a scene upside down vs. right side up scenes (figure 1). An inversion effect will show an “impairment” in the navigation system. This impairment should be seen when analyzing the accuracy of the subjects’ responses. We expect that categorizing and navigating in upright scenes will be matched in accuracy, while navigating in inverted scenes will be much worse than categorizing in inverted scenes (figure 2). Categorizing and navigating in upright scenes should be matched because, without restricting natural human processes, we should be able to navigate through a scene just as easily as they are able to categorize the scene. We expected an inversion effect in navigation because humans have never navigated through a scene upside down before and the brain is not “wired” that way, thus making it much harder than navigating upright. An inversion effect will provide the ground bases of evidence that will eventually allow me to move on to fMRI and show that there indeed are specific regions in the brain allocated towards either navigating or categorizing. Combined behavioral and neural evidence of such a “scene inversion effect for navigation,” but not categorization, will provide the strongest evidence to date that scene processing is indeed composed of distinct systems for navigation and categorization.
I found it surprising and infuriating how long the behavioral part of this experiment has taken me. It is weird to have worked so long in the forefront, creating and building the outline of the experiment, and not seeing the expected outcome. Science classes have taught me that by following a procedure you will get the correct result, but research has reminded me that science is about exploration and learning. Hopefully, I will be able to fix these problems and move on to fMRI and see if my hypothesis holds neurally as well as behaviorally.
Visit the Undergraduate Research Programs website to learn more about applying for Independent Research Grants.
Comments
Post a Comment