Multiple Traces of Dr. Doug Hintzman

Dr. Doug Hintzman is an emeritus professor at the University of Oregon and a critical scholar who developed ideas and research that fueled our understanding of cognitive processes over the past 40 years. I would include him as a prominent figure of the cognitive revolution (the field shifting movement that rebirthed the use of cognitive structures as psychological mechanisms and was the start of computational modeling of cognitive processes in the 1960s-1990s), but that would be unfair to him. Dr. Hintzman shaped cognitive understanding independently. Diligently. Whenever he found too many people working on the same topics as him, he moved on. His research may have moved Psychology and been incorporated into a shift in cognitive understanding, but he didn’t join any movement. That’s Dr. Hintzman. He does his research, is fascinated by testing and exploring the mechanisms of cognition, and, when things get crowded, he moves on. There are always more projects.

To start, Dr. Hintzman introduces himself:

Both of Doug’s parents were teachers and they moved around small towns in Montana, teaching in  Kalispell (pop. 22,000) and Custer (pop. <500). His dad was both a teacher and principle at his school. He describes them in this excerpt:

Doug was a good student, and would have taken courses with just about every teacher — including his parents — as part of the nature of being in a school of this size. Education was important in his family and Doug set off to go to college upon graduation. He attended Northwestern University (Evanston, IL) with an expectation to study something like architecture.

He describes some of that undergraduate experience, finding  his interests and getting engaged with Psychology, at the university in the following excerpt:

There are some great things in this excerpt! We start with Doug’s journey by train to Northwestern — that would have been days rolling from station to station to reach Illinois from Montana. He reaches Northwestern in 1959 and  starts his education with a general introduction to the social sciences. This broad approach is the same as described by so many others I’ve spoken with on this journey who experienced Psychology within a spectrum of social sciences, e.g., sociology, psychology, anthropology, political science.

Whatever he thought of Psychology entering his undergraduate degree, Doug soon experienced that very great  attraction for so many in the field: here is a way to scientifically study thought and behavior. It wasn’t just couches and talk therapy, but a chance to experiment, to study laws and mechanisms of the system. That concept excited Doug and he accepted a research position working in Dr. Ben Underwood‘s lab studying paired associations in memory (and, yes, that is an awful lot of foreshadowing in Doug’s early life for his eventual career studying memory). For those with an interest in the history of Psychology, Benton Underwood is  no slouch. He’s a National Academy of Sciences member who had a decorated career studying cognitive processes (including memory). As Dr. Hintzman recalled, Dr. Underwood had a well-funded, large scale lab with several  grad and undergrad students involved. And here’s the added curious note: these experiments that Doug ran were not particularly fun. They involved long lists of nonsense words for participants to learn, relearn, relearn again, and then, learn even more. The participants got bored. Doug got bored. The aim of this research was  for the participants to overlearn materials, which meant everyone was bound to get tired of seeing the word list. While the tasks weren’t necessarily engaging, being involved in the lab, meeting the grad students, seeing what one could do with experiments clearly left a positive impact on Doug.

He applied to grad school, an NSF fellowship to support that study (which was funded), and he headed back to the west. Doug Hintzman entered Stanford University to study…well…some area of Psychology. Something experimental, maybe something  physiological.

Dr. Hintzman describes getting started on his graduate work at Stanford:

Dr. J. Anthony Deutsch had a great career in Psychology and, for a time, he was at Stanford just as Doug began his graduate studies. Doug took to working with him immediately, but, as he describes in that excerpt, the experiment literally fizzled out. As he described, between the timing of Dr. Deutsch’s sabbatical and an unfortunate mishandling of the equipment that Doug had strung together, that experiment abruptly ended. So there was Doug. No mentor immediately available and the originally intended experiment (to directly stimulate rat brains and test the reward circuitry) was a no go. This was kind of a turning point for the start of Dr. Hintzman’s career. He had worked with Dr. Underwood at Northwestern, knew something  about memory, and needed a project. Without a project Doug would have faced expulsion from the grad program and maybe would have lost his military draft deferment (making him eligible for the war in Vietnam). Doug headed over to a young faculty member’s office to request to work on memory. That young faculty member was Dr. Gordon Bower (who eventually became of the most prominent memory researchers in Psychology’s history). At this point, in 1963-64, Dr. Bower was at the start of his career. He was ambitious, he was busy, and Doug had never met him. Doug introduced himself, Dr. Bower pointedly asked Doug not to waste his time, and they set off working together. For what it’s worth, neither has been big on wasting time in their careers.

Doug had some early interest in computing. He and some other graduate students took a programming course at Stanford. As noted at the start of this essay, the 1960s (and while Doug is a grad student) is the onset of the cognitive revolution. In this next excerpt, Dr. Hintzman describes his first attempt creating a computer simulation of a behavioral process.

That Simon and Feigenbaum tech paper opened a lot of people’s eyes to the possibilities of computing and both authors are seminal in computing and the creation of artificial intelligence. Herb Simon, in part from this early work, and later from his collaboration with Allen Newell was awarded a Nobel Prize in Economics in 1978 and in 1975 he and Newell jointly received a Turing Award. This tech report, a simulation model of serial learning, provided some inspiration to Doug. He saw what they had done, figured out what he could do with the Burroughs B5000 computer, and created a dissertation: a simulation of paired-associate learning. Think about those elements: paired associate learning is the topic of both Hintzman’s undergraduate work and now his graduate study; he has an understanding of what tools he might use from his interest and extra curricular enrollment in a computing course, and he receives this innovative article from Bower to refer him to the simulations of Feigenbaum and Simon.

Dr. Hintzman completed his dissertation in 1967 and began his career as an independent scholar.  He wasn’t particularly interested in doing simulation working or continuing with the computer modeling in his dissertation. He describes getting started on his lab at the University of Oregon, some years later (1969).

What interested Dr. Hintzman was the form of the memory. How was it being encoded? What was stored and how was that stored representation different than other memory traces? As he discusses in the excerpt above, Dr. Hintzman, his colleague at U of Oregon, Dr. Mike Posner was an early adopter of computers into his lab. Eventually, Dr. Hintzman did get involved in a memory simulation model and completed some highly influential and innovative work on Minerva 2.

He describes the creation and testing of Minerva 2 in the following excerpt from our conversation:

The student described, Genevieve Ludlam, was second author on a 1980 article they published in Memory & Cognition simulating an exemplar-based model. That simulation set of some of the parameters for Minerva 2. Minerva 2 simulated memory with parallel processes. That was fairly novel and critical — parallel processing is now the standard, but in the early days of memory models it was unclear and difficult to simulate the difference between comparing all exemplars with a search item or one at a time (in serial and not parallel). The early PC computers relied on a floating point processor, and thus were designed to process a series of items, adding/subtracting/multiplying, but rapidly doing each operation in sequence.  Parallel processing and the capacity to address all items in a set at the same time required a more complex approach. In addition, the features of an item in memory were stored in relationship to other features. If a feature is a color, red, then the model  would encode that not just as “red” but its similarity/difference to other features.  Hence, the color would be in the memory trace and, at its essence, it would be defined by how that color diverges from some features (e.g., blue) and similar to other features (e.g., orange).  Finally, that model encoded forgetting. In retrospect that is an obvious important aspect of memory, but in the early days simulations were highly constrained by the available computing power and the inclusion of forgetting was not obvious and not included in all models. Importantly, the model simulated a process of memory that entailed making multiple memory traces. This is not Dr. Hintzman’s current belief, but let’s explain the thinking at that time. What happens if you see a balloon? You create a memory trace. What if you see that balloon again?  Do you create a new memory? Change the old one? Well, in Minerva 2 and with some reasonable evidence, the theory was that each encounter produced a new memory trace. So a person might have dozens, hundreds, maybe thousands of memory traces of the same item, each for a different moment of encounter.  It allowed the person making that memory to not only remember something but to have a memory of remembering something that is different from another memory of that something. So I remember not just my son, but I remember him a million different ways, each somewhat unique with different clothes, times of day, interactions, etc.  The model was innovative, it challenged existing norms, and there were few memory simulations published.

Here Dr. Hintzman describes the context of presenting Minerva 2 to others in the field:

There are a few very prominent, very well-known computer simulation models from the early 1980s (including Hintzman’s Minerva 2). The University of Indiana, and specifically, Dr. Rich Shiffrin have also been pretty central to this area of Psychology.  Shiffrin and Hintzman met up at a conference and realize that they have developed similar projects. Shiffrin goes on to continue tweaking and creating SAM and other simulations. In contrast, Hintzman, as he described above, starts to work hard to disprove his model and move on to other interesting ideas! That’s Dr. Hintzman: the field started to get crowded and he saw greener pastures in other areas of memory research.

Dr. Hintzman reflected on multiple trace memory models over his career:

It’s funny, one of the first things I asked Dr. Hintzman was why he had retired. He still loves doing research, he is still highly influential, and he has great collaborators and colleagues at Oregon. He retired and gained emeritus status almost 20 years ago, and yet he is continuing to work.  He told me he wanted to be like a grad student: free to explore and to work on projects of his interest. He keeps his research projects small, he runs his own participants. His ideas have continued to evolve consistent with new data and new ways of approaching the critical concepts. Through the changes in his ideas,  he continues to highly value the phenomenological approach: what does he experience (or any of us) when we create a memory and later recall it? He still  values that experience and is still attracted to the means to scientifically study and understand it.

That’s a little from Dr. Hintzman, but there is much more to tell!

Here are some of the wonderful publications from Dr. Hintzman:

Hintzman, D. L. (1967). Articulatory coding in short-term memory. Journal of Memory and Language, 6(3), 312-316.

Hintzman, D. L. (1968). Explorations with a discrimination net model for paired-associate learning. Journal of Mathematical Psychology, 5(1), 123-162.

Hintzman, D. L. (1984). MINERVA 2: A simulation model of human memory. Behavior Research Methods, Instruments, & Computers, 16(2), 96-101.

Hintzman, D. L. (1986). ” Schema abstraction” in a multiple-trace memory model. Psychological Review, 93(4), 411-428.

Hintzman, D. L. (1988). Judgments of frequency and recognition memory in a multiple-trace memory model. Psychological Review, 95(4), 528-551.

Hintzman, D. L. (1993). 1 6 Twenty-five years of learning and memory: Was the Cognitive Revolution a mistake?. In D. E. Meyer & S. Kornblum (Eds.) Attention and performance XIV: Synergies in experimental psychology, artificial intelligence, and cognitive neuroscience (pp.359-392). Cambridge, MA, MIT Press.

Hintzman, D. L. (2004). Judgment of frequency versus recognition confidence: Repetition and recursive reminding. Memory & Cognition, 32(2), 336-350.

Hintzman, D. L. (2012). Memory from the outside, memory from the inside. In M. A. Gluck, J. R. Anderson, S. M. Kosslyn (Eds), Memory and mind (pp. 32-47). New York, NY:  Psychology Press.

Hintzman, D. L. (2011). Research strategy in the study of memory: Fads, fallacies, and the search for the “coordinates of truth”. Perspectives on Psychological Science, 6(3), 253-271.

Hintzman, D. L. (2016). Is memory organized by temporal contiguity?. Memory & Cognition, 44(3), 365-375.

Hintzman, D. L., & Block, R. A. (1971). Repetition and memory: Evidence for a multiple-trace hypothesis. Journal of Experimental Psychology, 88(3), 297-306.

Hintzman, D. L., Block, R. A., & Summers, J. J. (1973). Contextual associations and memory for serial position. Journal of Experimental Psychology, 97(2), 220-229.

Hintzman, D. L., Carre, F. A., Eskridge, V. L., Owens, A. M., Shaff, S. S., & Sparks, M. E. (1972). “Stroop” effect: Input or output phenomenon? Journal of Experimental Psychology, 95(2), 458-459.

Hintzman, D. L., Caulton, D. A., & Levitin, D. J. (1998). Retrieval dynamics in recognition and list discrimination: Further evidence of separate processes of familiarity and recall. Memory & Cognition, 26(3), 449-462.

Hintzman, D. L., & Curran, T. (1994). Retrieval dynamics of recognition and frequency judgments: evidence for separate processes and familiarity and recall. Journal of Memory and Language, 33(1), 1-18.

Hintzman, D. L., Curran, T., & Oppy, B. (1992). Effects of similarity and repetition on memory: Registration without learning?. Journal of Experimental Psychology: Learning, Memory, and Cognition, 18(4), 667-680.

Hintzman, D. L., O’Dell, C. S., & Arndt, D. R. (1981). Orientation in cognitive maps. Cognitive Psychology, 13(2), 149-206.

Hintzman, D. L., & Ludlam, G. (1980). Differential forgetting of prototypes and old instances: Simulation by an exemplar-based classification model. Memory & Cognition, 8(4), 378-382.

(Featured image is Dr. Doug Hintzman at his office and at the top of this post is an image of the Office of the Psychology Department at University of Oregon).