Education

1988 B.A. in Computer Science, University of Texas at Austin
1988 B.A. in Psychology, University of Texas at Austin
1991 M.A. in Psychology, University of Texas at Austin Thesis: Contrast Sensitivity Functions Measured During Dark and Light Adaptation
1994 Ph.D. in Psychology, University of Texas at Austin Dissertation: The Influence of Light and Contrast Adaptation Mechanisms in Spatial Vision

Professional Experience

Summers, 1984-88 Computer Programmer, Maxicom International, Richardson, Texas
Summer, 1989 Research Assistant, Neural Networks group of Texas Instruments, Dallas, Texas
Fall, 1989- Research Assistant with Dr. Wilson S. Geisler, Fall, 1993 University of Texas at Austin
Spring, 1994- Postdoctoral Fellow, University of Pennsylvania present

Awards

Summer, 1990 The John Wallace Dallenbach Fellowship University of Texas at Austin
Fall, 1992 - Spring, 1993 The Emma F. Clark Fellowship, University of Texas at Austin

Professional Activities and Organizations

Optical Society of America
Association for Research in Vision and Ophthalmology
Ad hoc reviewer for Vision Research

Teaching Experience

Teaching Assistant: The University of Texas at Austin, Fall 1988-Spring 1992
- Computer Methods in Statistics
- Computer Methods in Statistics - self-paced
- Methods in Experimental Psychology
- Sensation and Perception
- Cognition - undergraduate
- Advanced Cognition - graduate
Temporary Instructor: The University of Texas at Austin, Spring 1993
- Computer Methods in Statistics

Grants Funded

NIH postdoctoral grant #1F32EY06574-01 Title: Retinal Circuitry and Human Spatial Vision

Presentations at Scientific Meetings

Hahn, L. W. & Geisler, W. S. (1994). The Influence of Eye Movements and Local Adaptation on the CSF. Paper presented at the Annual Meeting of the Optical Society of America. Dallas, Texas.
Hahn, L. W. & Geisler, W. S. (1994). Local Light Adaptation Mechanisms in Grating Adaptation. Paper presented at the Annual Meeting of the Association for Research in Vision and Ophthalmology. Sarasota, Florida.
Hahn, L. W. & Geisler, W. S. (1992). Contrast Sensitivity Functions Measured During Early Dark Adaptation. Paper presented at the Annual Meeting of the Optical Society of America. Albuquerque, New Mexico.
Hahn, L. W. & Geisler, W. S. (1991). Contrast Sensitivity Functions Measured During Dark Adaptation and on Steady Backgrounds. Paper presented at the Annual Meeting of the Optical Society of America. San Jose, California.

Publications

Hahn, L. W. & Geisler, W. S. (1995). Adaptation Mechanisms in Spatial Vision I: Bleaches and Backgrounds. Vision Research 35(11): 1585-1594.

Manuscripts in Preparation

Hahn, L. W. & Geisler, W. S. The Influence of Light and Contrast Adaptation Mechanisms in Spatial Vision

References

On request.

Research Interests

January 1995

My broad research goal is to explain the way in which the human visual system transforms the images projected onto the two retinas into a cohesive visual percept. Two fundamental aspects of this transformation compose my main interest: spatial vision and adaptation. By understanding spatial vision one learns how information is derived from changes in light intensity across the retinal image. By understanding adaptation one learns how the visual system alters itself to become well-suited for the immediate environment. Rather than focusing solely on spatial vision or adaptation, my research examines the way in which adaptation changes spatial vision.

Adaptation is typically described as the way in which exposure to an adapting stimulus changes the perception of a target stimulus. In order to create a computational model of an adaptation process, at least two things need to be known about the adaptation mechanism: first, the nature of the signal producing the adaptation (for example, a light signal produces light adaptation) and second, a quantitative description of the way in which the adapting stimulus changes the perception of the target stimulus (for example, a reduction in the apparent intensity of a target signal by a multiplicative factor). This approach has been used to describe several light adaptation mechanisms and contrast adaptation mechanisms. I have used elaborations of this paradigm to distinguish between adaptation mechanisms and to make precise predictions about the way in which adaptation alters spatial vision in the dark and in an illuminated environment.

Past Research

The standard measure of spatial vision is the contrast sensitivity function which describes sensitivity to a sinusoidal modulation in light (a sinewave grating) as a function of the spatial frequency of the modulation (the number of cycles per degree of visual angle). The traditional contrast sensitivity function is band-pass in nature - sensitivity peaks at a spatial frequency of about 3 cycles per degree of visual angle and declines as the spatial frequency increases or decreases. The decrease in sensitivity as the spatial frequency of the target stimulus increases can be explained primarily as a property of the optics of the eye. The decrease in sensitivity as the spatial frequency of the target stimulus decreases is explainable only as a neural phenomenon. The textbook explanation of this latter neural phenomenon is that the desensitization is the product of a retinal neural interaction known as the center-surround antagonism. However, a center-surround antagonism cannot explain the absence of the low frequency decline in sensitivity for short stimulus durations. This failure of the conventional explanation led me to consider an unconventional hypothesis which attributes the desensitization at low spatial frequencies to an interaction between local light adaptation and eye-movements during steady fixation. (Local light adaptation is light adaptation which integrates over a small area of the retinal image.) This hypothesis can potentially explain the presence of the low frequency decline in sensitivity for long stimulus presentations and the absence of the low frequency decline for short stimulus presentations. A major goal of my dissertation was to test this alternative hypothesis.

To test this hypothesis I created a five phase plan: 1) collect data describing the time-course of local light adaptation, 2) collect data describing eye movements during steady fixation, 3) form an integrated model of both phenomena, 4) use the model to make explicit predictions about the way in which the interaction between light adaptation and eye movements would change the contrast sensitivity function and 5) test the predictions empirically. I developed a model which was completely defined by measurements of adaptation and eye-movements when the phenomena were independent, but I used the model to predict contrast sensitivity when the two phenomena interacted. The change in the contrast sensitivity function predicted by the model fit the measured change quite well. Thus, it is very likely that the decrease in sensitivity to low spatial frequency gratings with long durations is due to an interaction between local light adaptation and eye-movements during steady fixation. Although there certainly is a center-surround antagonism in the retina, it is not responsible for this decrease in sensitivity to low spatial frequencies. Because the contrast sensitivity function is widely used in describing the human visual system and in implementing computer-based visual systems, I believe that this new understanding of sensitivity to low spatial frequency stimuli will have a wide-spread impact on the field.

Another goal of my dissertation was to measure the cumulative effect of adaptation mechanisms on spatial vision. As mentioned earlier, one way of categorizing adaptation mechanisms is by the type of signal which produces the adaptation (i.e. light or contrast). However, a distinction can also be made between mechanisms which are quantitatively different, even though they may act upon the same signal. Two classes of adaptation mechanism which are characterized by their quantitative descriptions are subtractive adaptation, in which a portion of the adapting signal is subtracted from the target signal, and multiplicative adaptation, in which the target signal is divided by a portion of the adapting signal. These distinctions result in a set of four well-defined adaptation mechanisms: subtractive light adaptation, subtractive contrast adaptation, multiplicative light adaptation and multiplicative contrast adaptation.

Typically these four adaptation mechanisms are studied in isolation, but in the natural environment, these mechanisms almost certainly act in concert. One goal of my dissertation was to measure the coexisting influence of these mechanisms on spatial vision. I used a contrast matching task which allowed me to simultaneously determine the relative contributions of three of the four types of adaptation. After considering many plausible combinations of the four adaptation mechanisms, I concluded that the results were best described by a model which included only the multiplicative light adaptation mechanism. Although the simple model provided a good fit to the data, there were systematic differences between the data and the model which suggest other mechanism(s) are at work. Developing hypotheses about the missing component(s) in the model will be a goal of future research. Mechanisms which I plan pursue as possible components will include: luminance nonlinearities, cortical contrast nonlinearities and the division of stimulus parameters between different neural channels. All of these components can be expressed within the framework of a computational model.

In the work I have described so far the adapting stimulus has been a pattern - a sinewave grating. A sinewave grating can produce both light adaptation and contrast adaptation. Isolated light adaptation can be produced by changing the mean light level. This is the adaptation one experiences when leaving a dark movie theater for a sunlit parking lot. Classic measurements of the change in spatial vision produced by changes in the state of light adaptation have been made by measuring contrast sensitivity functions for different levels of background light. While the change in spatial vision as a function of background light level is well-documented, the change in spatial vision during dark adaptation is surprisingly undocumented. Furthermore, two existing hypotheses (a local multiplicative light adaptation hypothesis and an equivalent background hypothesis) make conflicting predictions about how spatial vision changes during dark adaptation. So, in a collaboration with Bill Geisler which preceded my dissertation (Hahn & Geisler, 1995), I resolved to measure contrast sensitivity as a function of time following exposure to an intense adapting field.

We traced the change in sensitivity of observers to a new spatial stimulus following exposure to a uniform field of intense light. (It was necessary to develop a new type of stimulus to make the measurements in the dark.) We also measured changes in the detectability of the new stimulus for different background light levels. This latter measurement allowed us to compare our measurements with other classic measurements to insure that the new stimulus was functionally similar to the traditional stimuli.

Our results showed that long-term dark adaptation produces the same increase in sensitivity at low spatial frequencies (coarse gratings) as it produces at high spatial frequencies (fine gratings). This finding supports the local multiplicative luminance adaptation hypothesis, and rejects the equivalent background hypothesis. In addition to discriminating between these two hypotheses, the data provide a singular description of how spatial vision changes during long-term dark adaptation.

Ongoing Research

My current work involves a hypothesized interaction between adaptation and spatial vision which was suggested by anatomical evidence. In recent years a conflict has arisen between the anatomical description of the retina and psychophysical measurements of human spatial acuity. Anatomists have found that cone photoreceptors are connected via electric synapses called gap junctions. The presence of gap junctions between cones suggests that the response of a single cone is not always independent of the response of adjacent cones. Psychophysicists, however, have measured the limits of spatial acuity at high light levels and determined that the visual system could reach these limits only if the response of each cone is independent of its neighbors. In collaboration with Peter Sterling, Andrew Hsu and Jack Nachmias, I have hypothesized that these gap junctions act as an adaptation mechanism. At low ambient light levels, open gap junctions might produce signal averaging among the cones which would enhance luminance signal detection at the cost of degraded spatial resolution. On the other hand, at high light levels when the luminance signal is strong, gap junctions might close in order to preserve high spatial sensitivity. Currently, as a postdoc at the University of Pennsylvania working in the Psychology department and the Neuroscience department, I am testing this hypothesis. (My funding is from the Institute for Research in Cognitive Science.) I am creating a functional (as opposed to a structural) model of the photoreceptor array and the gap junctions, and psychophysically measuring spatial acuity at low light levels.

Future Research

In the future, I plan to seek and establish hypothetical anatomical locations for psychophysically-described adaptation mechanisms. This will motivate the creation of new models which incorporate both structural attributes of the early visual system and psychophysically-measured functional attributes.

Finally, I also plan to test the limits of existing models of adaptation and spatial vision with more complex stimuli. Historically the literature has been dominated by a motivation to isolate and measure a single attribute of adaptation or spatial vision. This has meant using mathematically simple stimuli which are very artificial. I will test the models which predict response to simple stimuli with stimulus configurations which approach the complexity of stimuli found in the natural environment. The use of more complex stimuli is likely to introduce new types of adaptation mechanism, new types of spatial mechanisms, and new space-adaptation interactions. For instance, it is known that prolonged exposure to a drifting grating produces motion adaptation. Although some spatial attributes of motion adaptation are known, little is known about interactions between motion adaptation other types of adaptation. In cooperation with computational modeling, the study of visual sensation has produced a collection of functional building blocks which I intend to use to form more general models of visual perception.