Reports

HOPE Report No.37 2004-11-03

Report of Summer Internship for Foreign Students

mariko.yamaguchi

Period of Stay: May 31, 2004 - August 7, 2003

Introduction

 As a foreign student interns for the summer of 2004, I stayed at the Kyoto University Primate Research Institute in Inuyama, Japan, for 10 weeks. Under the guidance of Professor Tetsuro Matsuzawa in the Section of Learning and Intelligence, I participated in various activities during my internship and benefited greatly from the exposure to a new learning environment. My day consisted of assisting with several aspects of the lab, including feeding the chimpanzees, setting up, running, and cleaning up after the experimental sessions, and participating in numerous academic opportunities, such as the annual meeting of the Primate Society of Japan, guest lectures, and weekly seminars with the graduate students. It was through both of these settings that I was able to accomplish my goal of broadening my perspective and gaining experience as a scientific researcher.
Although I was at PRI as a foreign student intern, perhaps this title was not strictly appropriate; I had a second, non-academic goal in mind when I originally contacted Professor Matsuzawa about the possibility of spending the summer in his lab. I am Japanese by birth (and therefore not exactly "foreign") and am fluent in the language, but have spent the majority of my life outside of Japan. Thus, I am unaware of the implicit rules of Japanese society and the ways in which people interact with each other in the workplace. I had hoped to grow not only as a researcher but as a Japanese adult as well through daily contact with the professors, graduate students, and the many staff at PRI. During my ten weeks, I received the social education that I felt was necessary for myself.

The First Week: Observation

 My first week at the Primate Research Institute was spent observing various groups of researchers as they ran their own experiments with different chimpanzees. With 15 chimpanzees and 5 independent research groups, daily life at the institute escapes utter chaos only by running on a strict timetable. During the day, I rotated from group to group and observed the studies being run. Because I had no prior experience working with apes, and because chimpanzees are weary of strangers, even this minimal contact at the beginning caused some stress on both parties.
I was further frustrated by my inability to assist in the experiments; I was asked to merely sit and watch. Perhaps this was a difference in culture. Expecting to hear the clear directions to which I had become accustomed in the United States, I found it difficult to adjust to the silent requests and directions given by the professors and students at PRI and perhaps took longer than necessary to fully understand what was expected of me.
During this first week, I was able to observe all of the research groups and their studies. The professors, graduate students, and researchers were extremely helpful and patient in thoroughly answering my questions about their work, and I was exposed to new experimental paradigms as well as philosophies that differed from those I had known.

Assignment to a Research Group

 After this first week, I was assigned to work with Toyomi Matsuno, a graduate student working on several projects that explore the visual perception of chimpanzees, for the remainder of my internship. I assisted him with setting up the experimental booths before each session and in cleaning up afterwards, and I was also able to observe in detail the progress of his studies. I was somewhat familiar with one of his experiments that explored the perception of an ambiguous "collision" of two objects, having participated in several studies looking at human perception of similar events and through reports of a group testing capuchins (Cebus apella) and rhesus macaques (Macaca mulatta) under similar conditions at Yale University.
Because all of Matsuno-san's current experiments utilized learning paradigms where the chimpanzees must first learn to perform a certain task using the display on the touch screen, they all start with training at the very beginning. This task may be as simple as touching the larger bar or selecting a figure that matches the sample, and the chimpanzees learn what is required of them through trial and error. However, through further training on more complicated tasks and with combinations of different paradigms, the chimpanzees are able to participate in a large variety of studies. I had never seen this training process before and was fortunate to see how performance improved steadily.

An Original Experiment

 In parallel to assisting Matsuno-san with his experimental sessions, I began thinking about possible ideas for my own experiment. My background is mainly in infant cognition, where I have participated in projects exploring infants' understanding of animacy and intentionality, and I am also interested in object individuation and number comprehension. My first ideas were mainly based on previous studies performed with infants to test similar cognitive abilities in chimpanzees. However, infant experimental paradigms rely heavily on measuring looking time. Because the chimpanzees at PRI are not accustomed to merely looking at a display (either live-action or taped) while certain events take place, this paradigm could not be used.
In order to overcome this obstacle, I suggested altering the procedures to mimic current experiments being run, such as a match-to-sample task. For example, by seeing which conditions are more likely to be used by chimpanzees in selecting an "animate" being, as opposed to an inanimate one, the cues important to animacy can be found. However, a large portion of these studies required prior training and was unsuited for the short length of my stay.
I had never designed my own experiment prior to this and during this brainstorming process, which unfortunately occupied the majority of my internship, I sometimes felt uncertain of my ability to successfully design and execute a complete study. However, the graduate students and professors were extremely helpful in giving me valuable feedback. Matsuno-san and Professor Matuszawa were especially patient in listening to my ideas and suggesting alternatives when problems arose, and Professor Tomonaga and Professor Tanaka were kind as well in discussing their thoughts with me.
After much deliberation, I decided to explore the understanding of concepts in chimpanzees. By using a forced choice task in which the chimpanzees are simultaneously shown two different photographs and must select the image with an apple in it to be rewarded, I hoped to look at how different aspects of the stimuli influenced their understanding. In order to obtain the stimuli, I performed four separate searches in GoogleR images using these words: "apple," "grapefruit," "autumn(fall) leaves," and "car." I varied the language in which the search was run and also ran both the singular and plural forms. In the end, I selected 300 unique photographs of apples and 100 from the other categories, for a total of 600 pictures.
Next, I ran two separate tests with human subjects using the collected stimuli. One test asked the subject to rate the "apple-ness" of each stimuli (all stimuli, not just the apple photographs), and the other asked the subject to respond as fast as they could to the question of whether or not there was an apple in the displayed photograph. Both tests displayed the stimuli, one at a time, for a short interval on a computer monitor and the response was entered by a keyboard. The results from each test are attached at the end, separated into two graphs for each test: one for photographs of apples and another for other stimuli.
I had hoped for a correlation between the two results; that stimuli that were rated high on the "apple-ness" scale would elicit a quick "yes" response, and stimuli that were rated low on the scale would elicit a quick "no" response, with the middle area of the scale having a longer reaction time. However, there was no such correlation; the graphs (again separated into apple and non-apple categories) are attached at the end.
This can be attributed to the fact that it is likely these two tests are measuring different things. The yes/no reaction time has a small spread, indicating that there probably was a floor effect where the fastest amount of time it took for any response to be elicited overruled the variability in response time due to the actual stimuli. On the other hand, the rating measures how well the stimulus fits into the category of "apple."
Experimental sessions with the chimpanzees began as well. The stimuli were displayed in counter-balanced pairs, where each stimulus appeared an equal number of times within the session, paired with different photographs. The large number of stimuli prevented the chimpanzees from memorizing the correct photograph by rote. Because this was a learning paradigm, training had to be conducted first for the chimpanzees to learn to select the photograph with the apple in it; however, once they had understood the task, data could be analyzed about the correlation between the human ratings and the chimpanzee error rate for each stimuli, as well as the effect of the distracter stimuli. This experiment continues to be run, thanks to Matsuno-san. I hope to continue analyzing the data for any emerging trends.

Life with the Chimpanzees

 In addition to participating in the experimental sessions with the chimpanzees, I had the opportunity to take part in feeding them twice a day. This serves as an important time in fostering a friendship between the students and the chimpanzees, separate from an experimental setting.
Every morning, I prepared breakfast for Puchi and Pico, a mother and a 1-year-old infant who are currently isolated from the community because of medical issues. Professor Matsuzawa and I attempted to increase the amount of food intake for Pico by placing food directly into her mouth and trying various kinds of fruits. Although Pico was not doing well when I originally arrived at the institute and tended to spend most of her time clinging to her mother, she improved significantly with the help of blood transfusions and other treatments and was actively moving around the room, swinging from rope to rope using her arms, by the end of July.
In the evening, I prepared vegetables for dinner with the graduate students and helped feed as well. This was also an opportunity to get to know each individual better: not only what vegetables they liked and didn't like, but their separate personalities and quirks.
We also set aside two afternoons during my internship in which everybody - the professors, graduate students, researchers, and veterinary staff - worked to plant new trees in the chimpanzee enclosure. Although the majority of trees planted the first time was de-branched or uprooted by the next week, the second planting with more conifers fared better. This was one of the many attempts by the institute to ensure a high quality of life for the chimpanzees; the lessons I learned were not only about experimental science but about enrichment in animal care as well.

Other Activities at the Institute

 Although feeding and running the experimental sessions with the chimpanzees took up the vast majority of my day, typically from 8am until after 5pm, I had the chance to participate in several seminars and meetings.
Each Monday, there was a reading group with the graduate students where we rotated each week, selecting a chapter from The Cognitive Animal: Empirical and Theoretical Perspectives on Animal Cognition (eds. M. Bekoff, C.Allen, & G.M. Burhgardt. 2002, MIT Press.) and researched the reference materials and presented the chapter to the group. All discussions were conducted in English, and the goal was to provide an opportunity for the students to speak in English and to increase their scientific vocabulary.
There was also a weekly seminar in conjunction with the Section of Cognition and Learning, where researchers took turns presenting their current research. I was able to learn about the ongoing works of not only the professors and students at the institute but of the visiting professors as well. Of those, particularly interesting was the presentation by Dr. James Anderson about contagious yawning in chimpanzees, a study done with Dr. Matsuzawa and Dr. Masako Myowa-Yamakoshi, and by Dr. Shohei Takeda, about sand play in chimpanzees. I was originally unfamiliar with most of the work presented during these seminars and found the exchange of ideas both interesting and stimulating.
Once a month, the veterinary staff, the professors, graduate students, and other researchers met to discuss the community of chimpanzees. Any unusual events or medical issues concerning the chimpanzees were brought up at this meeting so that everybody was informed. In addition, everybody who worked with the chimpanzees recorded what they did, with which individual, at what time, in a shared notebook so that anybody interested could look up information about each chimpanzee, their health, their motivation level, and any other information of interest. I felt fortunate to be able to see all the inner workings of how this institute was run.
The annual meeting of the Primate Society of Japan was held in Inuyama during my internship, and I was able to attend the conference. Here, I was exposed to the rich history of Japanese primatology and met prominent researchers in the field, a privilege that I had not expected.
I also visited two psychology labs at the main campus of Kyoto University, seeing both the comparative lab and the developmental lab. I was able to talk with Dr. Kazuo Fujita and Dr. Shoji Itakura, hearing about their research as well as those of their graduate students.
All of these settings served as chances for me to encounter new researchers and new experiments, and I hope to keep in touch with these faculty and students who are currently or are sure to become significant contributors in the field of primatology.

Conclusions

 This internship was much more than I had hoped for, and for that, I must thank everybody at PRI. Professor Matsuzawa was patient and understanding, and always generous with his time, and Matsuno-san showed me the discipline necessary for a scientific researcher. In addition, this wonderful experience would not have been possible without the support of my advisors at Yale University, Dr. Laurie Santos and Dr. Valerie Kuhlmeier. I also cherish the connections that I made at and through the institute, and I hope to take further advantage of my background and contribute to research in both the U.S. and Japan, growing as a researcher.

Figures

HOPE　Project<hope@pri.kyoto-u.ac.jp>