Interactive exercise: Explanation
Groups 1 to 3 indicate the type of question you were asked. According to Collins and Quillian’s (1969) hierarchical network (see figure 7.3 in the textbook), the further the information relating to the object and the object are from each other the longer it should take you to answer the question. The questions in group 1 related to knowledge that was just one step away from the object. Group 2 related to information that was two steps away, and group 3, three steps away.
Collins and Quillian would predict that you should be quickest for questions from group 1 and slowest for those in group 3, with group 2 somewhere in between.
How did you get on? If your results didn’t match their prediction why might that be? Think about the role and effect of typicality in this task. “Is a penguin a bird?” is a harder question than “is a sparrow a bird?”. Why?
- Category-specific deficits:
- Disorders caused by brain damage in which semantic memory is disrupted for certain semantic categories.
- Ecological validity:
- The extent to which research findings (especially laboratory ones) can be generalized to everyday life.
- A type of schema in which information about objects and their properties is stored.
- A form of statistical analysis based on combining the findings from numerous studies on a given research topic.
- A term introduced by Bartlett to refer to the tendency in story recall to produce errors conforming to the rememberer’s cultural expectations.
- A type of schema relating to the typical sequences of events in various common situations (e.g. having a meal in a restaurant).
- Semantic dementia:
- A progressive neurodegenerative disorder characterized by gradual deterioration of semantic memory.
- Schemas incorporating oversimplified generalizations (often negative) about certain groups.
- Transcranial magnetic stimulation (TMS):
- A technique in which magnetic pulses briefly disrupt the functioning of a given brain area; administration of several pulses in rapid succession is known as repetitive transcranial stimulation (rTMS).
- Typicality effect:
- The finding that the time taken to decide a category member belongs to a category is less for typical than atypical members.
Research activity: semantic networks
Try creating your own personalized semantic network, just like in Figure 1. Simply start with an everyday object, say a toaster, and then work your way from the toaster with three or four semantic links; for example, toaster, kitchen, and silver. Then do the same for these semantic nodes. It is an interesting exercise and gives you some insight into the way that you personally store semantic information.
Figure 1:An example of a semantic network. From Collins and Loftus (1975). Copyright © American Psychological Association.
- Memory for lists: A demo based on Roediger and McDermott (1995).
- Dr Lawrence Barsalou talking about the brain’s modality-specific systems.
- Greenbert and Verfaellie (2010) discuss semantic and episodic memory in terms of research on brain-damaged patients.
Please find below biographies of two important researchers in the field of memory: Endel Tulving and Allan Collins.
The son of a judge living in Tartu, Estonia, Endel Tulving was born on May 26, 1927 and attended Hugo Treffner’s Gymnasium, where he developed his enthusiasm for sports, especially track and field. World War II drove the 17-year-old Tulving to Germany to finish schooling, beginning a separation from his parents that lasted 20 years. While Tulving was a stellar student, it was not until a year before graduating that he encountered a subject that truly interested him: psychology.
Prior to moving to Canada in 1949, Tulving spent time teaching, in addition to interpreting for the United States army and studying as a medical student at Heidelberg University for a single year. A year after his arrival, he married his wife, Ruth Mikkelsaar, and soon thereafter earned his B.A., which was immediately followed by a Master of Arts degree from the University of Toronto (1954) and his doctorate in experimental psychology at Harvard University (1957).
After Harvard, Tulving returned to what would become his primary academic home, the University of Toronto. Starting as a lecturer, he was appointed a University Professor in 1985, after having become a Fellow of the Royal Society of Canada in 1979. During his illustrious career, Tulving spent time teaching at Yale University and won a Guggenheim Fellowship in 1987, becoming a William James Fellow of the American Psychological Society in 1990. Two years later, he was named a Fellow of the Royal Society of London, later winning the Gairdner Foundation International Award for biology and medicine (2005), being made an Officer in the Order of Canada (2006), and being inducted into the Canadian Medical Hall of Fame (2007). Now an Emeritus professor at Toronto, Tulving continues to research at the Rotman Research Institute.
Tulving started off as a vision researcher, but, lacking the expensive equipment required to conduct such research, he quickly ventured into the field of memory when he became a lecturer at the University of Toronto. He initially sought to investigate the subjective organization of memory by asking participants to learn words and recall them in any order they chose, dissecting memories into those that are accessible given the current retrieval cues and those that are stored in memory, and thus available, but are temporarily unreachable. This work led to his discovery of the encoding specificity principle: the more similar the cues available at retrieval are to the conditions present at encoding, the more effective the cues will be.
Furthering his efforts to catalog and understand the various types of human memory systems, Tulving famously proposed the distinction between episodic (“remembering,” a uniquely human ability, according to Tulving; however, this has been subsequently questioned) and semantic (“knowing”) memories in 1972, representing a shift away from the standard, unitary theories of long-term memory. Episodic memories, with their autonoetic—or self-knowing—quality, afford the ability to mentally travel in time to recollect prior events.
In addition to purely behavioral research, including his Remember/Know procedure (1985), which sought to dissociate episodic and semantic memories on a subjective level, Tulving studied the deficits associated with different groups of amnesic patients and used neuroimaging techniques. Data from functional imaging studies gave rise to Tulving’s HERA (Hemispheric Encoding and Retrieval Asymmetry) hypothesis, which posits that the encoding of episodic memories involves the left frontal lobe whereas their retrieval depends on right frontal areas.
Tulving, E., & Osler, S. (1968). Effectiveness of retrieval cues in memory for words. Journal of Experimental Psychology, 77,593–601.
Tulving, E. (1972). Episodic and semantic memory. In E. Tulving and W. Donaldson (Eds.), Organization of memory (pp. 381–403). New York: Academic Press.
Tulving, E., Kapur, S., Craik, F. I. M., Moscovitch, M., & Houle, S. (1994). Hemispheric encoding/retrieval asymmetry in episodic memory: Positron emission tomography findings. Proceedings of the National Academy of Sciences USA, 91, 2016–2020.
Tulving, E. (2002). Episodic memory: From mind to brain. Annual Review of Psychology, 53, 1–25.
Allan Collins’ broad interests, cutting across the fields of artificial intelligence, semantic memory, and education are readily apparent in his professional pedigree. After receiving his B.B.A. in accounting in 1959, Allan Collins went on to pursue his Master’s degree in communication sciences, and then his Ph.D. in psychology, all at the University of Michigan. Before graduating with his doctorate in 1970, Collins began a long run as a Senior Scientist at the BBN Corporation, becoming Principal Scientist at BBN Technologies in 1982 and remaining there until 2000. Simultaneously, he was a professor of education and social policy at Northwestern University from 1989 until he retired in 2005.
During this time, Collins has also been associated with Boston College, as a Research Professor in the school of education between 1998 and 2002, as well as at Harvard University, first as a visiting scholar between 2001 and 2005, and then as a visiting senior lecturer for a year, beginning in 2005. Between 1991 and 1994, Collins co-directed the Center for Technology in Education with Jan Hawkins, evaluating the use of technology in educational settings.
Collins, the first chair of the Cognitive Science Society, serving from 1979–1980 is also a Fellow of the American Association for Artificial Intelligence, a Guggenheim and Sloan Fellow, and was editor of the journal Cognitive Science between 1976 and 1980.
Collins’ reaction time experiments, along with his colleague M.R. Quillian, at the BBN looked at how long it took individuals to make simple verification judgments about statements of fact. Their results yielded a model of semantic memory, for which Collins is most famous: The Collins & Quillian Semantic Network Model. In this model, the meanings of words are organized into a series of hierarchical networks, with the major concepts represented as nodes. Since information about the properties/features of different concepts is stored as high up the hierarchy as possible, redundancy is minimized. Thus, the model follows the rule of cognitive economy.
Problems with the model quickly became apparent, however. For instance, not all concepts are categorical and the time it took participants to verify statements was related not only to semantic distance, as Collins and Quillian suggested, but also was confounded with the frequency with which two words co-occur in language. Collins and Loftus responded to these criticisms in 1975 with a multidimensional spreading activation model that de-emphasized the hierarchical structure featured in the original model while emphasizing both the strength and distance that linked the nodes together. Not only did the new spreading activation model explain the typicality effect, it also predicted the phenomenon of semantic priming.
Much of Collins’ other work focuses on intelligent tutoring techniques, such as the SCHOLAR CAI system he developed with Jaime Carbonell and the WHY system. Collins has also been active in researching ways in which technology can be effectively utilized in academic and military training settings, with an eye for developing the ideal schools of the future and assessing their effectiveness.
Collins, A. M., & Quillian, M. R. (1969). Retrieval time from semantic memory. Journal of Verbal Learning and Verbal Behavior, 8(2), 240–248.
Collins, A. M., & Quillian, M. R. (1970). Does category size affect categorization time? Journal of Verbal Learning and Verbal Behavior, 9(4), 432–438.
Collins, A. M., & Loftus, E. F. (1975). A spreading activation theory of semantic processing. Psychological Review, 82, 407–428.