Student Research

Varun Wadia '15

Normally, steroid hormones have been understood to affect behavior by turning genes on and off. A simple illustration of this is puberty. A rise in steroid hormone levels in teenagers during puberty changes the way boys interact with and perceive girls. This method of action is very slow and typically takes a few hours or days in Goldfish but can take up to several years in other species (as is the case with teenagers and puberty). Recently, a rise in steroid hormone levels has been seen to change cell structure and behavior in as little as a few minutes in several species including rats, birds and fish1,2,3. This effect is too quick to be a genomic mechanism and is indeed unaffected by chemicals that prevent genes from being activated.

It is thought that high levels of hormone may be affecting sensory processing in Goldfish precipitating heightened behavioral responses to sensory stimuli. However, the external triggers that cause this rise in hormone levels are unknown. In other words, what cues in the mating environment cause the surge of internal hormone levels in the fish? What aspects of the opposite sex are they paying attention to? We do know that Goldfish can see the primary colors and also have a type of cone cell in their eyes that is maximally sensitive to ultraviolet (UV) light (356nm)4. Goldfish do not really change color with age or sex so it is thought that the UV reflectance patterns off their scales is what the fish use to distinguish one another and hence is what they would pay attention to in a reproductive situation.

My summer was spent exploring this idea. Old research in the Thompson lab showed that males were much more reflective than females hence we performed behavioral approach tests in which females are presented with a UV reflective male and a non-UV reflective male. The latter case is created using a plexiglass stimulus chamber that cuts off all UV light. Both the ambient light and chemical conditions were varied. We found that in an unaltered chemical environment, there is no significant preference shown regardless of light intensity. There is also no significant preference when the test fish were either injected with lutalyse (a chemical that induces high levels of circulating hormone – mimicking reproductive situations) prior to testing or exposed to the male pheromone Androstenidione (AD) during the test. However females that underwent both pre-test lutalyse injections and pheromone exposure during the test showed a significant preference for the UV reflective male at lower ambient light intensities. Goldfish are known to mate early in the morning under lower light intensities, thus these results suggest that UV reflectance patterns off the scales of sexually mature males may be an important visual cue for female mate choice in a natural setting.

However, ‘lower intensities’ simply implied a low power output setting on the bulbs used. In an attempt to be quantitatively rigorous about connecting the Goldfish’s behavior to the wavelengths they perceived, I worked with Prof. Msall of the Physics department to perform spectroscopy measurements on the light source at different power outputs. Spectroscopy will provide a comprehensive list of wavelengths present in the light and their relative intensities at a particular output setting. Performing these measurements on the bulbs we used to perform behavioral experiments will provide a much clearer understanding of the role of UV reflectances with regard to mate choice.

Unfortunately, I was unable to complete this part of the project in the allotted time as calibration of the equipment turned out to be quite a struggle. However I intend to complete the work in the fall and once complete we will be able to more rigorously corroborate or debunk the hypothesis that UV reflectance patterns are an important visual cue with regard to mate choice.

Ariye Krassner '14

My project was inspired by my plans to study abroad in Copenhagen, Denmark this past spring, as I anticipated working in a baby lab at the University of Copenhagen. During my time in Denmark, I was able to build professional relationships by assisting developmental researchers on their study of temperament, mother-child interaction and attachment, and began collaborating with them to a design a summer project that replicated some of their tasks in order to draw cross-cultural conclusions regarding early interaction patterns and emerging personality traits between Danish and American mothers and babies. In developing my protocol, I became particularly intrigued by the role of infants’ “surgency” (a broad dimension of temperament that includes tendencies toward high activity level, rapid approach, and the expression of positive affect, especially in arousing contexts) as an influence on relationship style and extended my research plan to include a thorough assessment of this temperament dimension.

The beginning of my summer was spent preparing a detailed protocol for the administration of eight tasks designed to assess different aspects of surgency, mothers’ levels of sensitivity and intrusiveness, general interaction patterns (e.g., maternal touching behaviors, reciprocity in parent- child vocal patterns) and the babies’ attachment orientation (secure or insecure). After the protocol was finalized and a detailed protocol was developed, I trained a research team to help collect and code the data, and recruited families to participate by visiting local child care centers, reviewing birth announcements in local papers and sending letters, and developing a website for interested parents. Thus far, we have conducted 38 hour-long visits and anticipate completing 6 more by the end of August. We have also begun coding child behaviors from each of our five temperament tasks. For example, for the “Arc of toys” task, a task that measures activity level and involves placing the child in the center of a standardized semi-circle of toys, my assistant and I have reached adequate levels of interrater reliability for our judgments of childrens’ intensity of movement, body position changes, and pace of play. For another task, the “Bubbles game”, which assesses the child’s levels of positive affect to a burst of soap bubbles, we are reliable for ratings of smiling, laughter, and vigor of approach.

I plan to follow-up this initial summer research in multiple ways. I have recruited a team of 4 students who will assist me in continuing to collect and code data this fall, with the goal of having complete data sets from at least 70 mother-child dyads by the end of the year. I will also be corresponding extensively with the Danish group to coordinate coding between our two labs to enable the cross-cultural comparisons. In fact, I have recently learned of the exciting possibility of one of my Danish colleagues traveling to Bowdoin in the fall to work closely with us on this endeavor. These data will form the basis of my honors project, to be completed by the end of the academic year. In addition, my advisor, Professor Putnam, and I have begun considering ways to organize the data that we have collected in order to produce between two and four submissions for next year’s meeting of the International Society on Infant Studies.

This project, which has immeasurably enhanced my academic growth and has the potential to contribute strongly to knowledge regarding the origins of personality, would not have been possible without the generous support of the Surdna Foundation. I am grateful for the opportunity they have allowed.