Monday, February 11, 2008

The Boyd / Osinga Round Table

Chicago Boyz' roundtable on Science, Strategy, and War: The Strategic Theory of John Boyd has finished. The entries are:



In academia, it's custom to publish these sorts of roundtables. Now the question is if Chicago Boyz will work with, say, Nimble to do the same here.

20:23 Posted in UNL / OODA | Permalink | Comments (9)

Monday, February 04, 2008

A History of the OODA Loop

This post was written as part of the roundtable on Frans Osinga's Science, Strategy, and War. Contributions have already been made by Chet Richards and Wilf Owen.

"To a certain extent the argument is valid that Boyd offered merely a synthesis of existing theories, a contemporary one, important and timely regarding the context of the 1970s and 1980s, but only a synthesis."
Osinga, 2007, pg 29


John Boyd's OODA Loop divides cognition into four processes, perception (called Observation), unconscious or implicit thought (called Orientation), conscious of explicit though (called Decision), and behavior (called Action). Frans Osinga's "Science, Strategy, and War: The Strategic Theory of John Boyd" does an excellent job describing the origins of Boyd's learning theory in the writings of Skinner, Piaget, and the cognitivists. However, Osinga's text excludes ongoing research into theories of learning related to OODA, as his text is focused on the development of the OODA model in particular rather than contemporary adaption. Fortunately, a recent review article by Jonathan St. B.T. Evans serves helps complete the picture, though the OODA loop is not mentioned there by name. Osinga's book is well worth purchasing, and can be thought of as as prolegomena to all future OODA work.


The OODA Loop


The "OODA loop," or "Boyd Cycle" (Osinga, page 2) is a dual-processing model of thought. That is, it supposes the existence of two seperate central executives inside each human mind. The first of these, "Orientation," is activated immediately by perception (called Observation by Boyd) and is capable of directly controlling behavior (likewise, called Action). Orientation is closely associated with long term memory. As Osinga writes on pages 236 to 237:

In order to avoid predictability and ensuring adaptability to a variety of challenges, it is essential to have a repertoire of orientation patterns and the ability to select the correct one according to the situation at hand while denying the opponent the latter capability. Moreover, Boyd emphasizes the capability to validate the schemata before and during operations and the capability to devise and incorporate new ones, if one is to survive in a rapidly changing environment.... verifying existing beliefs and expectations, and if necessary modifying these in a timely matter, is crucial. The way to play the game of interaction and isolation is to spontaneously generate new mental images that match up with an unfolding work of uncertainty and change, Boyd asserted..."


The second central executive, Decision, analogous to conscious thought, or what attention is spent on. As Osinga writes, "Decision is the component in which actors decide among actions alternatives that are generated in the orientation phase." Unlike orientation, decision faces limits in how much it can handle, and therefore relies on orientation to present it which simplified and categorized chunks in which to work.

John Boyd's model was purposefully designed as an cognitive and learning theory based on mainstream work within psychology. As Osinga writes on page 53:

On 15 October 1972 he wrote from his base in Thailand to his wife that ‘I may be on the trail of a theory of learning quite different and – it appears now more powerful than methods or theories currently in use’. Learning for him was synonymous for the process of creativity


In particular, Boyd's theory was based on the work of Jean Piaget, B.F. Skinner, and the earlier cognitivists. Boyd combined each of these traditions, though revised some elements. From Piaget he both took the concept of mental structures, as well as suspicion of the power of logical analysis alone as a proper epistemological tool. To again quote Osinga (page 68)

Boyd also came across another source of uncertainty. As Jean Piaget asserted in the book Boyd read for his essay, ‘In 1931 Kurt Gödel made a discovery which created a tremendous stir, because it undermined the then prevailing formalism, according to which mathematics was reducible to logic and logic could be exhaustively formalized. Gödel established definitely that the formalist program cannot be executed’.


As Osinga describes in Chapter 3, "Science," Boyd drew from both Skinner and the cognitivists the power of environmental feedback. Consider the relatively trivial cognitive or cybernetic proposition on page 72 that:

"A feedback loop is a circular arrangement of casually connected elements, in which an initial cause propagates around the links of the loop, so that each element has an effect on the next, until the last 'feeds back' into the first element of the cycle. The consequence is that the first link ('input') is affected by the last ('output'), which results in self-regulation fo the entire system.


Osinga then proceeds to discuss the OODA loop as Boyd applied it, touching only briefly on Chapter 7 of some applications of Boydian thought to areas of military operations. However, Osinga does not emphasize the areas in which the OODA loop itself is still unique, but only compares it to either incorrect renditions of the OODA model (such as the "simplified" rendition Osinga shows on page 2) or to theories that preceded OODA (such as a cybernetic model without feedback and "(Reflex)" instead of orientation or System 2, on page 75).

Consider, for instance, two other models, one by Jon St. Evans published in 2006 and the other by Richard Moreno, published in 1990. Using different terms, the Evans model describes the role of Orientation (called by him System 1) and Decision (called by him System 2). Orientation or System 1 initially activates, and it may either lead to conceptual change or else inform further System 2 deliberation. However, Evans' model lacks the cybernetic or cognitive function of feedback, and does not describe how the last function would inform the first. Boyd's OODA loop, by attaching both Action and Observation to the environment, therefore may be described as a completed Evans model.


Information Processing


Likewise, the OODA loop completes the Moreno model. Moreno's description of learning focuses on the transformation of information in the external world to long term memory. In particular, Moreno's ongoing research focuses on the limited ability of explicit though to handle all information that should be learned. However, Moreno does not view long term memory as much other than an end-state for information (rather than the abode of Boyd's Orientation or Evans' System 1). Additionally, like Evans, Moreno does not connect the last stage of his model with his first.


Dual Processing


Just as Osinga does not compare the OODA loop with other contemporary models, he does not describe contemporary research that further describes the difference between Orientation and Decision. The research on the subject is now well established, and Table 2 in Evans' 2008 paper "Dual-Processing Accounts of Reasoning, Judgment, and Social Cognition," in the 2008 edition fo the Annual Review of Psychology, provides a synopsis of the distinction between Orientation (System 1) and Decision (System 2)

System 1System 2
Cluster 1 (Consciousness)
Unconscious (preconscious)Conscious
ImplicitExplicit
AutomaticControlled
Low effortHigh effort
RapidSlow
High capacityLow capacity
Default processInhibitory
Holistic, perceptualAnalytic, reflective
Cluster 2 (Evolution)
Evolutionarily oldEvolutionarily recent
Evolutionary rationalityIndividual rationality
Shared with animalsUniquely human
Non-verbalLinked to language
Modular cognitionFluid intelligence
Cluster 3 (Functional characteristics)
AssociativeRule-based
Domain-specificDomain-general
ContextualisedAbstract
PragmaticLogical
ParallelSequential
StereotypicalEgalitarian
Cluster 4 (Individual differences)
UniversalHeritable
Independent of general intelligenceLinked to general intelligence
Independent of working memoryLimited by working memory capacity


Frans Osinga's Science, Strategy, and War is a groundbreaking book on the OODA loop, describing in excellent detail how it originated. Buy it. What is needed now is an comparison of the OODA loop to contemporary theories of learning and an application of OODA in light of the newest research.

References
Evans, J. St. B. (2006). The heuristic-analytic theory of reasoning: Extension and evaluation. Psychonomic Bulletin & Review, 13(3), 378-395.
Evans, J. St. B. (2008). Dual-processing accounts of reasoning, judgment and social cognition. Annual Review of Psychology, 59, doi:10.1146/annurev.psych.59.103006.093629.
Mayer, R.E. (1996). Learners as information processors: Legacies and limitations of Educational Psychology's second metaphor. Educational Psychologist, 31(3/4), 151-161.
Osinga, F.P.B. (2007). Science, strategy, and war: The strategic theory of John Boyd. New York: Routledge.

11:20 Posted in UNL / OODA | Permalink | Comments (1)

Monday, November 12, 2007

Survey closed (Now for some words on Creativity and the OODA loop)

Thanks to all who participated in the recent study on creativity and blogging.

The pseudo-experiment was part of my larger study of the OODA loop applied to education, in this case to creativity.

I have only started to analyze the data. Before I launched this survey, I assumed that affective "gut" attitudes would be a better predictor than cognitive "thoughtful" attitudes, with respect to blogging. I also assumed that behaviors that signify purposeful practice on blogging would be directly related to recognition as a blogger, as measured through Technorati rank. Both these findings appear to be supported by the data.

As part of my dissertation proposal, I will suggest a second, larger test (using a different sample pool), to provide further support.

Over the next few days and weeks, I will post more description and analysis of the data. Eventually what I write here will find its way into my dissertation proposal (hopefully!). But for now,

Thank you.

You made this possible.

Saturday, November 03, 2007

Automaticity (Automation of Schemata)

Curtis was kind enough to except some of my thoughts on automaticity over at Dreaming 5GW. The section he highlighted was from OODA Alpha, which is an early draft of an academic paper I am writing on John Boyd's OODA (Observe-Orient-Decide-Act) loop, applied to educational psychology. Since then, much of the paper has been rewritten. Below are the paragraphs from the current draft that reference automaticity. While the the rest of this post is assembled from different sections of my OODA paper, they serve to put my current thoughts on automaticity, with respect to the latest research. So without further ado, some words on automaticity, or the automation of mental structures:

Within the OODA model, two aspects of decision are most apparent. First, it is slower. While orientation can directly guide action, decision making represents an additional step to this process. However, just as harmful is the dampening of the power of orientation and its related automaticity. Decision can be the distracting result of environmental conditions that make exploiting one's prior knowledge impossible. A fuller discussion on this phenomenon of disorientation is included below.

One of the sub-processes of orientation is analysis / synthesis, so it is no surprise that System 1 and System 2 activities together lead to a change of orientation. An example of this dual-system approach to learning can be seen in Leahy & Sweller's investigation into cognitive load theory (Leahy & Sweller, 2005). After defining learning as the creation and automation of the appropriate schemata, the authors conduct two experiments into this reorientation. Their finding is that in processing complex information, directing experienced learners to imagine a correct answer produces better learning than directing them to study the appropriate material. That is, deciding to rely on orientation produces a good reorientation.

The ultimate result of reorientation is automated, effortless, orientation-level control over tasks. This result, and how to get it, is aptly described by Topping, Samuels, & Paul (2007). In experiments focused on improving reading comprehension, both the quantity and quality of practice is shown to matter. Again the interaction between orientation and decision in reorientation is revealed. With each new quality practice, the mental schemata related to a task are reinforced, requiring less decision to execute them. At the beginning of learning, large-scale decisive control over action is required. At the end, decision does not impact the result as actions are implicitly guided and controlled from orientation.

Read more ...

11:12 Posted in UNL / OODA | Permalink | Comments (1) | Tags: automaticity

Wednesday, October 24, 2007

OODA Alpha, Part XIII: Conclusion

The Observe-Orient-Decide-Act, or OODA, loop is a model of human cognition. The OODA model is a dual processing theory that has two main circuits: Observe-Orient-Decide which is analogous to Level 1 processing, and Observe-Orient-Decide-Act which is a form of Level 2 processing. Within an educational context, one central insight of the OODA model is that an educator does not have to focus on decision, or conscious processing, to change actions. Two broad methods, reorientation and disorientation, are presented that operate by modifying or disrupting mental cognitive structures.



Three broad educational contexts are described. Instruction, or educating to some specific end, academics, or learner interaction supervised by an educator, and creativity, or the process of an educator preparing a learner to create new and useful products. For instruction and creativity, educators must focus on building the correct orientations within learners so they can learn. For academics, educators should use disorientation where appropriate in order to interrupt the natural behavior of learners to manipulate peer interaction. For creativity, educators should reorient learners so they possess the proper intrinsic motivation to be both well adjusted and successful.




OODA Alpha, a tdaxp series
1. Abstract
2. Dual Processing Systems
3. The OODA Loop
4. Decision
5. Orientation
6. A Theory of Mind
7. Reorientation
8. Disorientation
9. Education
10. Instruction
11. Student Interaction
12. Creativity
13. Conclusion
14. Bibliography

07:15 Posted in UNL / OODA | Permalink | Comments (0)

Tuesday, October 23, 2007

OODA Alpha, Part XII: Creativity

Science advances. While a literature on creativity exists in the OODA program of research (Boyd, 1976b), it draws on the conception of creativity as a fundamentally different form of thinking (Osinga, 2007). Modern research is converging on the realization that whatever creativity is, it is not the result of processes that are different than other forms of thinking (Kalyuga, Chandler, & Sweller, 1998; Kalyuga, et al., 2003; Kalyuga & Sweller, 2005; van Merrienboer & Sweller, 2005; Weisberg 1986, 1993, 2006;). Therefore, the antiquated sources of the original OODA paradigm (Osinga, 2007, 79) are set aside and modern research on creativity is examined in light of the observation-orientation-decision-action learning cycle.



Creativity is an understudied field within educational psychology (Plucker, Beghetto, & Dow, 2004). Creativity is defined as production that is “novel and interesting and valuable” (Simon, 2001) and is essentially an unstructured social process between individuals and already acknowledged experts in a field (Csikszentmihalyi, 1996). That is, creativity is not seen merely as divergent thinking , which may well be part of a special potential for creativity, (Torrance, 1968; 1993; Plucker, 1999) or only useful for studying the past (Simonton, 1984) which is certainly part of creativity, but rather the production of the novel, the interesting, the valuable whenever and wherever it occurs as long as it is recognized by an appropriate audience.

Read more ...

09:25 Posted in UNL / OODA | Permalink | Comments (2)

Monday, October 22, 2007

OODA Alpha, Part XI: Student Interaction

Human nature makes no sense except in the context of social interaction (Vygotsky, 1978; Boyd, 1986; Tooby & Cosmides, 1992; 2005; Bjorklund & Pellegrini, 2002; Alford & Hibbing, 2004). This allows the academic system to function. Whether learners are engaged in construction rationality through multiple perspectives and reflection (Piaget, 1932/1965, 1985, 2001; Moshman, 1995, 2005) or peer tutoring (Topping & Bryce, 2004), academics cooperation is used in a variety of contexts (Das & Das, 2004; Ping & Swe, 2004; Carter & Hughes, 2005; Nambissan, 2005).



However, while cooperation may come naturally and easily from learner's orientation, it may not be the form of cooperation that educators wish. Various forms of fairness driven cooperation needs to be suppressed by teachers such as cheating (Lin & Wen, 2007) and classroom disruption (Paulsel & Chory-Assad, 2005). An approach to modify cooperation among learners is needed, so that it can be disrupted where it hurts and encouraged where it helps.

Read more ...

09:20 Posted in UNL / OODA | Permalink | Comments (0)

Sunday, October 21, 2007

OODA Alpha, Part X: Instruction

In the context of conflict, Boyd (1986) instructs the reader to “get inside the adversary observation-orientation-decision-action loops (at all levels by being more subtle, more indistinct, more irregular, and quicker – yet appear to be otherwise.” The requirement for deception is because operations on the other's orientation take place in confusing, disordered, and menacing environment (Boyd, 1986, 5). Yet whatever the attrition educators face (Smith & Ingersoll, 2004), students are not a force to be taken-down but one to be built-up. Therefore, the proper application of OODA theory would be to get inside the learners' OODA loops, to manipulate their orientation bypass their decision making in order to change their implicit knowledge (Osinga, 2007). An instructor knowledgeable informed by OODA theory seeks to reorient learners.



In learning new words, to use as an example that teaches useful information, instructors wish to present an example of a new word in use, display what the word symbolizes, and explicitly define the word (Stahl, 1986) though too much information impedes performance (Igo, Kiewra, & Bruning, 2004; Igo, et al., 2007). That is, instruction is harmed by disorientation, even when the information that's impeded cognition would by itself be useful (Kalyuga, Chandler, & Sweller, 2000, McCrudden, et al, 2004). Likewise, techniques that avoid disorientation by emphasizing materials for learners improve instruction (Kiewra, 1985; Titsworth, 2004; Titsworth & Kiewra, 2004 Neef, McCord, & Ferreri, 2006), beecause cues requires learners to make less decisions.

Read more ...

09:20 Posted in UNL / OODA | Permalink | Comments (1)

Saturday, October 20, 2007

OODA Alpha, Part IX: Education

Twice, Boyd (1986) includes headings that read “? Raises Nagging Question ?” (27, 71). These headings are used to draw attention to a problem drawn by above statements. In the same way, the discussions above on orientation and decision, and the meta-cognitive processes of reorientation and disorientation, raises the nagging question how to apply the OODA loop to education. Clearly, some aspects of the OODA loops of a student population are beyond control. Managed heterosis in order to improve genetic heritage would may only a slight improvement in performance (Mingroni, 2007) at a socially unacceptable cost (Graves, 2001), while informed social engineering (Skinner, 1976) has produced only mixed results (Kinkade, 1973; Kuhlmann, 2005). Similarly, learners come from a whole range of prior experiences (). In the next section, therefore, applications of the OODA loop are considered which speak to var only new information. Indeed, the presentation of new information intended to influence learning is the definition of teaching (Eisner, 1964).



The next three sections are organized in order of decreasing educator control. First, the instructional environment, where an educator interacts directly with a student, is examined. Second, the academic environment, where learners interact with each other under rules devised by the educator, is explored. Finally, the creative environment, where the only educators are those of the field in a domain that a learner chooses to engage, is discussed. The focus also shifts from the need to implicitly alter orientation on the spot, to the requirement to selectively disorient, and finally the need for long-term improvement in learner orientation. Therefore, the following section can be viewed as the journey from reorientation to disorientation, and back again.




OODA Alpha, a tdaxp series
1. Abstract
2. Dual Processing Systems
3. The OODA Loop
4. Decision
5. Orientation
6. A Theory of Mind
7. Reorientation
8. Disorientation
9. Education
10. Instruction
11. Student Interaction
12. Creativity
13. Conclusion
14. Bibliography

09:15 Posted in UNL / OODA | Permalink | Comments (0)

Friday, October 19, 2007

OODA Alpha, Part VIII: Disorientation

Disorientation, defined as "Mismatch between events one (seemingly) observes or anticipates and events (or efforts) he must react and adapt to" (Boyd, 1986, 115) is in cognitive terms the process of removing a situation from Level 1, or orientation-state, control and deliver it over to the more error-prone. Disorientation is not a process of the OODA loop itself, but rather a meta-process than can be done to alter the operation of the OODA loops of others.



Disorientation differs from concepts familiar to educational psychology in that disorientation, as originally envisaged, was designed to degrade the performance of a thinking enemy (Luttwak, 1987) to the point of his defeat (Hart, 1991; Danchev, 1999). Boyd (1986)'s description of disorientation was to “Uncover, create, and exploit many vulnerabilities and weaknesses, hence many opportunities, to pull adversary apart and isolate remnants for mop-up or absorption” (117). However, educational psychologists already know about disorientation and use it as a tool under a different name: cognitive load (Chandler & Sweller, 1991). Either too much or too little cognitive load degrades performance and leads to disorientation (Teigen, 1994; Paas, Renkl, & Sweller, 2004). Too much cognitive load prevents Decision, forcing a reliance on Orientation (Bargh, 2000). Cognitive load is measurable (Brunken, et al., 2003; Paas, et al., 2003; van Gerven, et al, 2004) and so disorientation is open to scientific manipulation.

Read more ...

09:10 Posted in UNL / OODA | Permalink | Comments (0)

1 2 Next