International Conference of the Learning Science
ICLS 2002, Seattle
Jay L. Lemke
Univeristy of Michigan
jaylemke AT umich DOT edu
Abstract:
Why do some learning experiences contribute to sustained long-term patterns of action and personal development, while the effects of other experiences are short-term and soon extinguished? How do we learn to flexibly sustain long-term projects and agendas of inquiry? How can we validly assess students’ learning of intellectual practices which cannot be carried out on the timescale of typical tests?
I will present a view of sustained learning as an aspect of development that takes place through the emergent integration of processes on multiple timescales. If we envision learners as participants in complex social-ecological systems, we can begin to understand why large-scale social and cultural factors, including diversity, play a key role in learning that lasts. We can also begin to restore respectability to the role of affective, emotional factors in a comprehensive view of identity, learning, and development.
How does what we learn in minutes add up to what we learn in a lifetime?
What keeps learning alive over the longer timescales of our lives?
What makes learning last?
I want to propose today that we consider adding these important questions to our research agenda in the Learning Sciences.
At various times in the history of the field it has been useful to make a separation, conceptually and methodologically, between learning and development. I believe that many of the critical issues of interest today in our work require us to re-examine more closely the sense in which learning that lasts is an aspect of longer-term developmental processes.
What are the multiple timescales of learning and development, and how are they integrated with one another, in our lives and in our approaches to curricula and teaching?
At every moment of learning we are already also engaged in on-going activities and agendas on many longer timescales:
The scale of learning episodes that may last several minutes or more,
of classrooms periods on the timescale of about an hour,
of the educational day across very different activities and subjects,
of the social day that extends into the hallways, home, and streets;
of the agendas of months and years in the formation of our social identities, the development of relationships with others, working for our survival and for our future.
What we learn in a few minutes may contribute to learning that lasts a lifetime or, more often, it may not.
Likewise what we learn in a curriculum unit of weeks, or in a course of months.
How much of what we learn do we forget hours later, days later, months later, years later?
And what do we remember? What do we continue to think about, make use of?
And WHY?
Early research on very long-term memory, from Ebbinghaus onwards, has provided us with some interesting facts and clues to the larger puzzle I would like us to consider.
There are also other conceptual frameworks for examining these issues, for example by asking about how artifacts in the material environment and institutionalized practices in various social settings help us to remember by affording us opportunities for the renewal and further extension of what we have learned.
Before I present some details about these various considerations, I want first to say something about why I think these issues are important.
I am not just interested in promoting long-term memory. I am interested a key implication of examining the integration of learning across multiple timescales:
Learning that lasts because it has become a part of who we are, because it has become part of our habits of thinking and working, is learning that finds a home in the larger-scale and longer-term projects of our lives. This means that if we are concerned about learning for life, and not merely about learning for examinations, we need to look beyond the very short timescales of lessons and the school day to the activities and agendas that occupy students across their days and across the multiple settings, outside school as well as inside it, where they are making their lives, deciding who they are and who they want to become.
What are the aims of education, and indeed of specific curricula in science learning or mathematics or literacy learning, that cannot in principle be assessed on the short timescales of test items or even in-school essay writing, because the practices and habits to be learned take much longer times to be enacted?
If we want students to be able to design an experiment, explore the implications of a system of axioms, critique and synthesize viewpoints of diverse authors, manage their own independent learning, successfully undertake collaborative group inquiry …
Then we know that the necessary practices which enact and demonstrate this learning take days or weeks or months, not minutes or even an hour.
How do our curricula and teaching sustain the support needed to foster these learning aims over long time periods, which may be an order of magnitude longer for learning than for enactment … more than a year to learn and practice, perhaps several weeks to perform and demonstrate?
How does our assessment system provide for the measurement of these important learning outcomes?
And above all, how much attention are we paying to students’ other agendas-in-progress on the timescales of such longer-term learning?
Why have we allowed our schools and curricula to become so isolated from students’ lives and learning outside school? Why do we take so little account of their home lives, their time spent with peers, the impact of mass media and popular culture – all that they live and learn, usually with more attention and engagement, and often with more long-lasting effects, than what they learn in our classrooms?
There is a serious danger in American education today (and through its influence, in global education as well) that our vision of education is becoming too narrowly technical. Are we prepared to abandon the traditional higher aims of education just because they cannot be tested in an hour or two? Are we content to be educators who focus on the learning of specialized information and principles, with no thought given to the human development of our students, whether they are becoming the kinds of people we will be proud of? The kinds of people who will make wise choices about the future of society and this planet?
Where are the great lessons of life learned? Are they learned in classrooms? Are they learned through purely cognitive activity, regardless of affect and personal engagement? I think we all know that there is deeply meaningful learning, learning that lasts, which takes place in our life experience outside of classrooms. Can it be that learning in classrooms need take no account of this greater learning outside? Can we expect that academic learning which does not take account of students’ developing personal, social, and cultural identities, which does not take account of their learning from life outside the classroom can possible last? Can possibly become a part of who they are, how they think, what they decide?
I am not here to make a purely philosophical argument. I want to ground an approach to these issues in concrete research and specific theoretical developments. But I also want to urge us to renew the connection between what we really believe is important in people’s lives and the work we do in the learning sciences. Not just for moral or social reasons, but for very precise scientific reasons.
The scientific study of long term learning, at least as regards memory, began with the classic studies of Hermann Ebbinghaus on the “curve of forgetting”, which showed how much of what we learn is lost and how soon.
Later follow-ups to this line of research shifted from the classic nonsense syllable memory of early research to re-testing of students on curricular content, months or years after their initial successful learning.
Short-term to sustained vs. extinguished learning: Ebbinghaus’ Curve of Forgetting: without review, after 30 days, only 20% is retained; in later studies of student classroom learning this often dropped to as low as 2-3% retained;
Review of studies of very long-term retention:
“knowledge of experimental design
and statistics also showed no significant decline over the whole of the 12 year
retention period, suggesting that active use of this knowledge in practical work
made it more memorable, and possibly reflecting a superior retention of
procedural, as opposed to declarative, knowledge.” (Cohen et al. 1992)
When grades were decomposed into
examination marks and coursework marks it was found that:
- high coursework marks were associated with good retention but
- examination marks were unrelated to retention.
[Very Long Term Retention of Knowledge. Gillian Cohen, Martin Conway and Nicola Stanhop. March 1992. http://kmi.open.ac.uk/publications/techreports.html]
What we know, coming out of this work, is basically that we forget most of the specifics and details we learn very quickly, even if we learned something with high emotional engagement, and that what we remember much later on is what has been reinforced by regular use and renewal. This is more often procedural knowledge and general principles than factual information.
What evokes use and renewal? We can influence our own memory deliberately by coaching ourselves to regularly review and renew, but the natural process is the one by which the environment regularly and repeatedly evokes the need to know and use some element of our knowledge. We not only learn by doing, we remember by continuing to do.
Even from research on autobiography and its reliability as a historical account, we know that our view of our own past is the one we rehearse to ourselves, in stories we tell, in stories we hear. We remake our lives retrospectively as coherent narratives, and in the process we distort and replace actual memories of the more chaotic and ambiguous lives we really led.
The social situations that evoke telling stories about ourselves are akin to the many social and institutional sites and contexts in which we continue to use what we have learned, repeatedly.
What then happens to learning when we change classes every hour and teachers every few months?
What sustains the kinds of learning habits, the sorts of cognitive and meta-cognitive skills, the habits of strategic and critical reasoning over the longer timescales needed for them to become part of who we are for the long term?
Contrary to many early hopes, we also know that these wonderful abstract habits of mind do not exist for the most part in the abstract. They are very much bound to specific concrete contexts. We reason in a particular way in relation to a particular content. We articulate and argue in a particular way in a particular community. Take away our relationship with a particular teacher, or our participation in a particular group of peers, and we lose part of who we were in the process of becoming. Long term identities and intellectual habits do eventually become part of who we are and remain, but how long does that take?
How long should a student maintain a relationship with a significant teacher in order for the learning which is embedded in their relationship to mature and become self-sustaining as part of the student’s long-term identity?
How long should project work continue, or working groups maintain their cohesion, so that the learning, particularly the higher-order learning, the meta-level learning, which is embedded in them can become learning that lasts?
How do we assess higher-order learning longitudinally over appropriate timescales, probably multi-year timescales? What assessment technologies do we have for this purpose?
What are the multiple timescales of learning and development? What are the timescales of the significant processes: cognitive, interactional, curricular, social, biographical processes by which the learning of minutes or hours adds up to the learning of years and lifetimes?
See TIMESCALES CHART
To what extent do our core theoretical notions require re-definition or re-interpretation at different timescales?
Does “identity” mean the same thing at the timescale of identity assumed in a group interaction on the timescale of an hour or an hour a day spread over a several weeks, as it does if we are talking about the “identity” we take on in a long-term personal relationship with a teacher over several years? Or the “identity” we have as a member of a culture and society over a significant fraction of a lifetime?
Does “learning” mean the same thing on these different timescales? Does “development”? Is the problem of “retention” or “transfer” the same?
Are the process which contribute to change in typical patterns of behavior the same when we are speaking about changes from minute to minute and changes from month to month or year to year or decade to decade?
In recent years complex systems theory has attracted the attention of researchers in many fields, in part precisely because it deals with issues of multiple timescales in the dynamics of natural and social systems.
One promising model for human social-institutional systems, our social ecologies, are complex biological systems, such as organisms and ecosystems. In such systems we need to examine what is happening at any one level of organization, which is the focus of our attention, not only in its own terms and on its own spatial and time scales, but we also need to study its place in a series of larger and smaller units.
We can study the behavior of a cell in its interactions with other cells, but not apart from analyzing the behavior of cell membranes and other constituents at a much smaller scale, and not apart from studying the larger tissue structures and organ functions in which the cell and its interactions with its fellows participates.
We can study the discourse and behavior of students in interaction with one another and a teacher across a classroom episode in its own terms, but not usefully unless we also look at the meanings of individual utterances in terms of their lexical content and grammatical forms, and not to any point unless we also look at how discourse and behavior in an episode is co-determined by the place of that episode in the development of larger themes and aims in a lesson and a curriculum.
In these terms we can imagine learning as an emergent phenomenon: new organization of behaviors which pre-existed at N-1 level, subject to constraints from N+1 level (and heterochronously N+n levels mediated through semiotic artifacts) … the re-signification and over time the alteration of characteristics of the N-1 behavior patterns as they are selectively strengthened by their participation in longer-term sustainable emergent level-N patterns.
In most physical and chemical systems, and many biological systems, these issues of multiple timescale analysis are simplified by the Adiabatic Principle: significant amount of energy are not efficiently exchanged between processes operating at radically different rates or timescales … and where information exchange is based on direct energy transfer, we usually need to only consider the two scales immediately adjacent to the one that interests us at the moment.
But in many human social-ecological systems, information which influences behavior, can be transmitted across very long timescales, compared to those of action itself, by such means as written texts, architectural plans, diaries, and in general by material artifacts that carry significant cultural meaning: artifacts that can be read as signs.
In more mundane classroom terms, these artifacts that promote integration across timescales can include items on a bulletin board, the period table of the chemical elements on the wall, the arrangement of chairs in the room, the textbook itself, a persistent and re-used measurement instrument or set of manipulatives. It can be a website to which successive classes over several years have been making contributions.
If the usual problem of cross-scale integration in complex systems is reducible to analyzing a nested hierarchy of levels of organization, looking at sets of levels only three at a time (N-1, N, N+1), the problem for systems with semiotics, i.e. with transmissible sign-based meanings that go beyond the direct physical interaction effects of an object, are far more heterarchical. Influence travels along networks of interconnectedness of social practices, mediated by artifacts, such that events remote in time and space become relevant to the here and now in ways that would be very unusual in other sorts of complex systems. This phenomena has been termed “heterochrony”.
An important practical implication of such an analysis of the dynamics of social organizational systems and learning is that the processes and networks which operate across longer timescales tend also to operate in more spatially extended systems.
If we are interested in what a student is thinking and why he or she is or is not learning, paying attention, interested, motivated, engaged, reflective, making connections to this or that … then the material system in which the relevant longer timescale processes are taking place is not usually confined to the space of the classroom.
We have evolved a methodology in classroom studies of returning day after day and with our video cameras to capture the longitudinal processes of learning and development within the same classroom. A great improvement over older studies that looked only at single lessons, or as I did in the 1980s at sequences of three lessons. But the students do not vanish into the vacuum like so many virtual tachyons between Monday’s math class and Tuesday’s. They go down the hall to an English or a Science class. They go to lunch with their friends. They go home. They watch television. They may surf the net, do online chat, play music, watch a DVD movie, take recreational drugs, go dancing, have sex, get assaulted, witness a violent crime, go shopping.
All in the course of 24 hours. How much in the course of a school year?
Educational ethnographers have begun to confront these issues. They follow students home, and to the mall and the movies. They are starting to pose questions about the disconnect between students experiences in classrooms and in the rest of their complex lives. About the disconnect between the curriculum and what happens outside the school.
Many of the longer timescale processes in which students participate take place along the trajectory of the personal lives, not within the artificial institutional boundaries of school courses and class periods. They pass through out classes, but they were somewhere else only minutes before and will be elsewhere minutes and hours after, for all the time before we see them again in the same classroom.
Those longer-term processes are having effects in our classrooms that we are not aware of. They are relevant to learning in ways our curricula totally ignore.
If it is true that learning which lasts does so because it is sustained and renewed and evoked by the environment on intermediate timescales between the timescale of initial learning and the timescales of integration into long-term development and identity, then it is those aspects of learning which are ALIKE between the classroom and students’ out-of-class experiences which are going to be most sustained and supported.
What do we teach students that continues to be evoked and sustained in the hallways, at lunch, at home, among friends, in participation in popular culture and mass media? What do we know about this?
We can consider this issue to be in some ways the inverse to the classic problem of transfer of training. Traditionally researchers in the learning sciences asked how what was learned in school could be transferred and applied to everyday life. What this view of complex ecosocial systems and multiple timescales for learning proposes is that we ask how the everyday lives of diverse students maintain and support some abstract aspects of what we teach in school? Which aspects? For which students? How? For how long? With what effects on long-term learning and development? With what consequences for the organization of learning in the classroom, and for approaches to education that do not conceive themselves in a purely endo-scholastic way, but imagine learning as a process that takes place across human trajectories which weave in and out of schools and other institutions and settings on multiple timescales.
There is another theoretical perspective on learning and development which also frames these questions.
The educational sociologist Pierre Bourdieu developed over a period of many years the theory of social habitus. In essence Bourdieu’s research, across many domains of human activity, including education, shows that people develop long-term dispositions to respond to situations, objects, people, opportunities, etc. in very specific but mostly unconscious ways which they tend to share with others who have had the same sorts of social backgrounds and histories.
Bourdieu attempts to bridge between the categories of the cultural and the biological. He proposes that social habitus is embodied in us in very physical ways, in the way that a trained athlete responds almost automatically to a fast changing situation on the field, or a trained musician lets a new score flow through his fingers and his years of training to produce a more fluid and stylistically interpreted performance than any novice would make.
The source of habitus for Bourdieu is learning, whether in school or elsewhere, but his special contribution is to point out that such embodied habits of mind are specific to our culture, to our period of history, to our generation, to our social class, to our place in and our trajectory through the diversity of a complex modern society.
His work can be read as telling us once again that the kinds of learning that last are those which are characteristic of larger units of organization than the classroom, the lesson, the day, or the school. They are characteristic of the social milieu in which we lead the significant parts of our lives, and schooling matters within this mix just to the extent that it is reinforced by our experiences in the rest of our lives.
Of course our postmodern lives are complicated. There are contradictions today for so many people between different aspects and moments of their lives. We are not bound to remain in the milieux in which we begin our lives. We often cross boundaries to pass through different communities, with different values. We mix our participation in aspects of the wider culture and its diverse subcultures in ways that allow us to build multiplex identities and find reinforced in ourselves different aspects of the habits of mind formed in the home, the school, our many peer groups, our many institutional commitments.
Sometimes comfortably, sometimes very uncomfortably.
Which brings me to the final point I wish to make. A very large part of learning is how we feel about what we are learning, how we are learning, who we are learning from, and who we are becoming in the process.
I believe that our field has been postponing the task of coming to grips with the affective components of learning long enough. We are in danger of narrowing our very definition of learning to a purely cognitive process and permanently casting out the essential and critical affective dimension of the learning process.
Emotion or feeling/affect is an aspect of an over-time bodily process, in which we are always also further participating in some larger environmental process, and all these processes occur on and across particular timescales.
Emotion or feeling in this view is produced by the unfolding, enactment, performance of these body-in-environment processes. We can therefore view emotional feelings in part as indices or signs of the nature of the processes, including social practices, that we are engaged in.
If we ask ourselves questions such as:
How do we feel about what we are doing?
What is this activity doing to us, what is it making us feel?
Then we realize that there is affective knowledge here about the interaction of the body/self and the doing of these practices-in-context. It is the kind of knowledge that clinical psychologists attend to in themselves, and that for example E. Gendlin’s “focusing” practice teaches people to better attend to [www.focusing.org]. It provides useful knowledge about the nature of our participation in various processes/practices, and about processes or practices that may be to some extent inhibited or blocked, or simply unable to complete, to carry forward toward subsequent processes.
The carrying-forward notion in phenomenological psychology links process on one timescale to its integration into processes on longer timescales, and can connect not just a series of processes at a given timescale to one another, but also to the longer-term issues and agendas which are relevant to whether they carry-forward or not.
How long does it take to change how we feel about science? Or mathematics? About ourselves? About more particular or specific practices?
The timescales for affect and affect change are key kinds of knowledge relevant to the processes of learning and development, and to their connections with social-institutional change (in both directions).
Feelings and emotions also ramify across multiple timescales. Some feelings are specific to brief periods of time. Others are long sustained. Many of the dispositions we consider important for academic learning: rationality, dispassionateness, objectivity, calm … are themselves affective states or affective aspects of processes. We are all aware of the contradictions between enthusiasm and the joy of learning and the call for more dispassionate stances in learning.
We are also all aware of the extent to which anger, anxiety, frustration, depression, fear, and other emotions play a significant role in academic learning, and not learning, for too many students.
We are well aware of the importance of emotional bonds between teachers and students, both positive ones and negative ones, and the contradictions of professionalization, which calls for impersonality in student-teacher relationships, and the fundamental facts about the role of emotionally significant relationships in learning.
There is far more about all this that we do not know. The present emphasis on the concept of identity development as a key element that education must take into account is once again opening the door for research on affect, feeling, and emotion as essential elements in understanding learning and human development.
It is time to bite this bullet. It is time to go against the long cultural traditions of upper-middle class, north European, and particularly masculine academic ideologies which disparage emotion as unworthy of study, mystify it as incapable of being studied, and above all associate it with the childlike, the feminine, the working class, and the invidiously denigrated cultures of the world.
How long have students been telling us that the schooled approach to education is boring? That they miss the human dimension in the teaching of science and mathematics?
How long have we known the magical effects of humor in the classroom?
How often have we recognized significant breakthroughs in learning and in classroom ethos by their emotional signs?
How long are we going to allow commercial mass culture to feed the emotional hunger of our students in order to increase profits at the price of shallowness of both thinking and feeling? And not respond to this need ourselves with an approach to learning that understands and makes good use of its affective dimensions?
Analyzing learning across multiple timescales in a social-ecological perspective leads us outward from learning events along the longer-term traversals of our lives through the diverse places where we make fully-felt meanings to complex learning that lasts.