What are intentions, and how are they related to consciousness? In the late 1800’s, Franz Brentano argued the two were part of the same thing. That is, he claimed that consciousness was intentional (Ash, 1995). What he meant by this was that consciousness always seemed to be directed toward something, be it an object in the environment or an object of thought. Intentionality therefore, according to Brentano, referred to the object-directed nature of consciousness.
Nowadays, the word intention is most-often used in cognitive scienceto refer to the thoughts one has before producing an action (Wegner, 2002). For example, before you actually started reading this article, you may have thought to yourself, “Consciousness on the edge...huh…sounds interesting, maybe I should read this.” This interpretation of the term intention is consistent with the way we use the word in a legal context. The difference between murder and manslaughter is not whether the defendant’s actions caused another’s death, but whether the defendant intended (i.e., thought about or planned) the death.
Here then are two definitions of the term intention. While the former, older definition treats it as an inherent aspect of consciousness, the latter, more-modern definition treats it as a conscious precursor to action. In what follows I describe how I think the term underwent this change, and then attempt to make a case for reviving the former in favor of the latter.
A Brief History of Intentionality and Consciousness
At the turn of the last century, the notion of intentional consciousness was laid aside as psychologists migrated toward the stimulus-response methodology of behaviorism. Instead of disappearing completely however, consciousness and intentionality manifested themselves within different names in different fields. Intentionality and its connotation of directedness was conceptualized as residing in behavior, and was implied via terms such as drive reduction. Consciousness, on the other hand, and its denotation of first-person phenomenology, found conceptual shelter in the term perception and a research home in the field of psychophysics.
Figure 1. A brief history of intentionality and consciousness in Experimental Psychology.
Underlying the success of this conceptual framework was its general acceptance by both behaviorists and psychophysicists. Behaviorists studied the lawful relationship between environmental events (i.e., stimuli) and behavioral events (i.e., response), while psychophysicists investigated the lawful relationships between environmental events (i.e., stimuli) and perception (i.e., consciousness). As a result, consciousness came to be thought of as experiencing, and was localized functionally on the sensory input side of the organism. Intentionality, on the other hand, was associated with doing, and was localized functionally on the motor output side of the organism. Consciousness and intentionality therefore were functionally segregated.
With the advent of cognitive psychology, cognitive psychologists began to once again appeal to internal states and the ability to plan. Consciousness and intentionality however, were still, for the most part, associated with experiencing and doing, respectively. To be sure, early cognitive psychologists did discover that consciousness (i.e., perception) could be pre-tuned (Broadbent, 1958; Cherry, 1953). But instead of referring to such pre-tuned experience as intentional consciousness, researchers referred to it as attention-attenuated perception. Attention therefore, referred to the pre-tuning of perception, and intention, the pre-tuning of action. Later cognitive psychologists claimed such attentional pre-tuning was done for the sake of real-time behavior production (Allport, 1987), and thus coined the phrase attention for action. This terminology of course, reveals a rather strong commitment to the idea that attention has to do with experiencing while intentions have to do with doing. Within cognitive science therefore, consciousness and intentionality still remain divided.
Perception and Action-Planning: A Common Function
Despite the historical persistence of this division, more and more data are coming to light that seem to be forcing a re-evaluation. Neurophysiological and cognitive behavioral data for example, indicate that perception and planning actually seem to share common processing resources. Mirror neurons (Rizzolatti, Fadiga, Fagassi & Gallese, 2002) for example, seem to be involved in both the detection (i.e. perception) and pre-specification (i.e., planning) of agent-object relationships (i.e., tool use). Rizzolatti et al. conceptualize mirror neurons as resonators. What is gained by the use of the term resonator is the fact that the activity of mirror neurons entails both seeing and planning content. That is, as they resonate to the activities of conspecifics, the resonation simultaneously serves as detection (i.e., perception) and planning (i.e., pre-specification). In short, at least as far as some parts of the brain are concerned, there seems to be no strong functional separation between consciousness (i.e., seeing) and intentionality (i.e., planning). (Also see Decety, 2002.).
As regards cognitive-behavioral data, researchers at the Max Planck Institute for Psychological Research in Munich have been investigating the relationship between perception and planning for some time. The preeminent theoretical framework for this research is the Theory of Common Coding (Hommel, Müsseler, Aschersleben & Prinz, 2001). This theory asserts the following: (1) actions are planned in terms of the distal effects they are to produce, and (2) perception and action-planning share resources.
In a recent application of this theory, Kerzel, Jordan and Müsseler (2001) asked subjects to indicate the vanishing point of a moving stimulus. When subjects were allowed to track the moving stimulus via eye movements, the perceived vanishing point was displaced beyond that actual vanishing point by an amount that varied directly with stimulus velocity. When subjects were asked to fixate on a stationary target during the presentation of the moving stimulus, there was no forward displacement of the perceived vanishing point. To determine whether this differences was due to the presence of the eye movements (i.e., action) or the plan to move the eyes (i.e., intention) Jordan, Stork, Knuf, Kerzel and Müsseler (2002) required all subjects to fixate on a centrally located fixation cross and indicate the perceived vanishing point of a stimulus that moved on a circular trajectory around the fixation cross. In one condition the offset of the moving stimulus was produced by the computer (i.e., the Induction condition). In another, subjects were asked to make the stimulus vanish by pressing a button (i.e., the Intention condition). In another, they were asked to press the button (which made the dot vanish) as soon as the dot began to move (i.e., the Cue condition). The point of the manipulation was to have subjects do basically the same thing (i.e., press a button while fixating) for different reasons (i.e., to make the dot vanish, or in response to the dot’s onset). In the Induction condition, the perceived vanishing point was actually behind the actual vanishing point. In the Intention and Cue conditions the perceived vanishing point was displaced toward the location specified in the subjects’ intention (i.e., towards the initial position and final position in the Cue and Intention conditions, respectively). Collectively, these cognitive-behavioral data are consistent with the notion that perception and action planning share resources. Thus, getting ahead or falling behind in conscious space seems to be a matter of the types of events one is planning. If one’s plans are at the anticipatory edge of the present (e.g., tracking a moving target with the eyes) the anticipatory planning required of such continuous movement creates an anticipatory context for spatial consciousness, due to the common resources of planning and perception. If one’s plans refer to past events however (e.g., press a button after the stimulus appears) phenomenal space becomes somewhat reactionary.
Consciousness, Intentionality and Event-Control
Given the degree of convergence between neurophysiological and cognitive-behavioral data, it may be time for cognitive science to move beyond models of consciousness that partition it from planning and action. Planning and seeing seem to constitute a common function, what I have referred to as a distal-event system (Jordan, in pressaa,b). To reconcile this notion with traditional models, I have developed the model illustrated in Figure 2, what I term the event-control model. In this model, the traditional psychological functions of perception, planning (i.e., cognition) and action are modeled as synergistically coupled control loops, each of which controls events at different spatio-temporal scales. Within such an event-control hierarchy, the distal-event system specifies a distal effect (i.e., organism-environment coordination) to be achieved. Lower level control systems (i.e., action systems) then control the limb movements necessary to attain the specified distal effect. Given the coupling between these systems, changes in body-environment relationships produced by the action control systems feed back into the distal event system (what we traditionally refer to as perception) and, as a consequence, transform the moment-to-moment content of the distal event system. In short, one has two-nested control systems that are synergistically coupled yet function at two different spatio-temporal scales of control.
Figure 2. The Event-control Hierarchy.
The additional loop in Figure 2 (i.e., the virtual-event loop) is meant to model the human ability to generate and evaluate virtual events (i.e., thoughts) that can ultimately constrain the events the lower level loops work to control. As an example, one can generate and evaluate a plan to catch a flight to the Bahamas. Up until the moment one is actually seated in the plane, the event ‘catch a flight to the Bahamas’ is a virtual event. Regardless of its ‘virtual’ status however, the plan is fully capable of constraining one toward producing events in one’s immediate environment such as buying tickets, packing clothes, and driving to the airport. And once one is engaged in producing such events, these plans (e.g., ‘buy a ticket’) are fully capable of constraining one’s immediate relationship to the environment. Figure 2 then, replaces the traditional single-loop model of psychological functionality (i.e., Figure 1) with a hierarchy of nested control systems, each of which is responsible for pre-specifying, monitoring, and producing outcomes at a different level of scale.
Given that these systems are all coupled and function simultaneously, one can be said to be controlling (i.e., pre-specifying, monitoring, and producing) multiple events at the same time. The temptation of course is to identify consciousness with one particular level of event control. It seems to be the case however that the actual timbre of consciousness can find itself distributed across different levels of event-control at any given moment. A beginning pianist for example, is probably very conscious of his/her attempt to control the relationship between finger positions and keys. Pre-specifying, monitoring, and producing these relationships constitutes the task at hand. For an expert pianist however, who has long since automated the control of finger-key relationships, consciousness may be most intimately tied to the control of the emotional states provoked by the piece. Thus, the content of consciousness is fluid and dynamic, and pinning it down to one particular level of event control seems difficult, if not impossible. One can dance in the dark (i.e., action-event control), one can dance across a well-lit room (i.e., distal-event control), and one can dance across a well-let room while thinking about how nice it would be to be dancing with another (i.e., virtual-event control). In all three cases, the intentionality (i.e., directedness) and consciousness (i.e., phenomenology) are contextualized by the constellation of events one is working to control. If this is the case however, then experimental psychology is to some extent, back where it started, for intentionality once again refers to the directedness of consciousness. This time around however, it is not a naked phenomenal directedness that is divorced from action. Rather, within the event-control model, intentionality and consciousness constitute a gradient (as opposed to a dichotomy) whose context is determined by the distality of the events a given system works to control.