Flow

From LearnLab
Jump to: navigation, search

Flow is the affective state of optimal experience that creates pleasure by balancing the challenge of the task at hand to the skills of the person. Originally discovered and popularized [1] by Mihaly Csikszentmihalyi ("chick-sent-me-high-ee"), flow has been studied across diverse tasks ranging from athletics to learning by students to games. More recently, flow states of students using learning systems have been measured to understand its relation to learning outcomes.

Foundations of flow

Flow, as originally formulated by Csikszentmihalyi and described in his book [1], has 8 components that are highly associated with the state of flow:

1. A challenging activity that requires skill

Flow is associated with challenging activities that requires skill to accomplish. In particular, flow occurs when the "opportunities for action perceived by the individual are equal to his or her capabilities." [1](p52). An activity too challenging leads to anxiety; not challenging enough leads to boredom.

2. The merging of action and awareness

Flow activities are associated with the complete occupation of attentional resources by the task at hand.

3. Clear goals and feedback

Individuals need clear, specific goals and feedback moment to moment on if they are being accomplished. For example, a painter might have criteria to decide, for each brush stroke, if it is good or bad.

4. Concentration on the task at hand

Individuals completely ignore any task irrelevant stimuli or recollections from memory.

5. The Paradox of control

Flow activities are associated with the perception of control rather than the actuality of control. Individuals perceive that their actions have consequences: if they act just right, they can achieve perfection. But external events may always cause this control to be imperfect.

6. The loss of self consciousness

Individuals no longer attend to their ego or engage in self scrutiny or judgments of self.

7. The transformation of time

Due to the intense concentration on the task, perceived time is no longer related to actual time.

8. The Autotelic experience

Flow activities are intrinsically rewarding - doing the activity itself is the reward, not reinforcement after the task (e.g., money). Individuals engage in flow activities for the sake of engaging in flow activities.


The data for results come from two types of studies. Csikszentmihalyi interviewed a huge variety of individuals and asked them to report on their most enjoyable experiences. From these retrospective accounts, he pulled out components and aspects that frequently occurred. Csikszentmihalyi also employed the experience sampling methodology in which participants were equipped with a beeper which semi-randomly went off. This triggered the participant to complete a short survey reporting on aspects of their current activity and current affect. These studies had the benefit of not being as clouded by memory biases or generalizations.

Flow in education

Flow has been used as a construct for understanding relationships between interventions and outcomes for students in the classroom. In particular, in a study examining the conditions in which after school activities improve students' rates of absence, suspension, and lateness and their English grades, experiencing flow during these activities made students significantly more likely to show positive outcomes [6]. Thus, flow is a useful measure of the utility of after school activities in benefiting students.

Flow in learning systems

More recently, studies investigating the affective state of learners using computerized learning systems and intelligent tutors have measured flow alongside other affects such as boredom, confusion, and frustration. Flow is particularly important because it is a desirable state for learners to be in. Flow has been shown to be associated with learning gains. In a study of students using Autotutor, flow was found to have a correlation with learning gains (knowledge pre and post task) of .29 [2].

However, attaining the benefits of flow requires careful design of the system to ensure students achieve flow. In a study of a hypertext learning system, students using an improved version of the system were not more likely to experience flow [8]. The authors speculate this failure could be due to the high incidence of apathy, which might have been caused by the rigid standardization of the tasks, reduced challenge, and reduced perceived skills. But flow was worth achieving in that, across all dimensions, the enjoyment and quality of the experience was highest in flow states.

Temporal dynamics of flow in learning systems

Flow does not necessarily persist indefinitely - it requires the right conditions to remain present. When learners were asked to report their current level of challenge and skills after each 7 subtasks using a learning system, whether or not students were in the flow zone - optimal balance between skill and challenge - fluctuated wildly with students jumping between flow, boredom, and anxiety [7]. Interestingly, these measures of flow over the course of the activities were unrelated to a final post-task assessment of enjoyment and control.

Theories of learning have postulated that the transitions between affective states are important for realizing learning gains. To empirically examine such theories, studies have examined transitions between affective states in the course of using a learning system [4][5]. Flow has been found to be a sink state - learners in flow are more likely than chance to stay in flow rather than transition to other affects. And they are unlikely to transition to boredom, confusion, or delight. Consistent with the idea that flow is associated with concentration on the task at hand, they are also less likely to game the system.

Measuring flow

Depending on the setting and purpose, a variety of approaches have been used to measure flow. In interviews, flow has been measured by its correspondence to the components identified by Csikszentmihalyi [1]. In experience sampling method studies, these measures have typically been less direct and focused on enjoyment of the activity rather than individual components.

There are several approaches to measuring flow in lab or field studies of learners using learning systems. One approach is to ask participants to rate their flow experiences. Studies have used methods such as rating the challenge and skills available at different points in the task [7]. Another is to have experimenters observe participants' actions and facial expressions to rate affective state, including flow [4][5]. This is usually done by sampling over a period of time to give the observer enough data to make a rating or allow them to switch between multiple participants.

To enable learning systems to use affective ratings of participants to influence their behavior, completely automated measurements of flow are desirable. To do so, both vision sensors and posture sensors recording the pressure on the seat at different points have been used. One study found posture sensors to be the most informative for detecting flow [3].

References

1. Csikszentmihalyi, M. (1990). Flow - the psychology of optimal experience. New York, Harper.

2. Craig, S.D. (2004). Affect and learning: an exploratory look into the role of affect in learning with AutoTutor.

3. D'Mello, S., Picard, R., and Graesser, A. (2007). Towards an affect-sensitive AutoTutor. IEEE Intelligent Systems.

4. Baker, R.S.J.d., Rodrigo, M.M.T., and Xolocotzin, U.E. (2007). The dynamics of affective transitions in simulation problem-solving environments. Affective Computing and Intelligent Interaction. 666-677.

5. D'Mello, S., Taylor, R.S., and Graesser, A. (2006). Monitoring affective trajectories during complex learning.

6. Shernoff, D. J., Vandell, D. L, & Bolt, D. M. (2008). Experiences and emotions as mediators in the relationship between after-school program participation and developmental outcomes. Long-term impact and outcomes of out- of-school time programs, Symposium conducted at the annual meeting of the American Educational Research Association, New York, NY.

7. Pearce, J. M., Ainley, M., & Howard, S. (2005). The ebb and flow of online learning. Computers in Human Behavior, 21, 745–771.

8. Konradt, U., Filip, R., and Hoffmann, S. (2003). Flow experience and positive affect during hypermedia learning. British Journal of Educational Technology, 34(3), 309-327.