Are you confused? February 4, 2019

What do we mean when we say we’re confused? We certainly know what it’s like to be or “feel” confused. We may be struggling with instructions on how to assemble something or trying with some difficulty to understand an issue in a textbook or wondering which turn to take to get to a destination. The circumstances under which we say we’re confused obviously vary widely, but there must be some common features that lend the term everyday utility in talking with others.

But first…

The reason for putting “feeling” in quotes was explained in my last post, “When are feelings really feelings?” (November 2018). In brief, the post argued that the sense in which behavior analysts discuss feelings concerns actual physiological events that the individual can detect and eventually learn to label through instruction by the verbal community – that is, sensations that can actually be felt. However, the unavoidable difficulty of teaching someone to label private events to which the trainer does not have direct access encourages a broader and potentially misleading use of the term. We often refer to feelings under circumstances in which it is doubtful that we can detect any real sensations. Instead, we tend to casually insert the term into everyday dialect with the vague implication that the referenced “feeling” in some way helps explain our behavior. This implication suggests an internal process of some undefined sort, though absent the requirement for any detectable physical foundation.

The distinction between these two ways of talking about how we feel is important. In one case, feelings are tied to real physiological events that can in principle be directly investigated, just as we can also study how the verbal community teaches us to describe them. In the other, no such physiological basis is required, leaving only the option of studying the environmental sources of control over these verbal referents, as well as addressing gratuitous implications that they reflect inner sources of control.

Should we consider confusion as a physiological state that the individual can detect? This is a difficult question for characteristically cautious scientists because there may always be physiological correlates to the ebb and flow of behavior under changing environmental circumstances. The issue, however, is not whether there are distinctive physiological correlates to our moments of apparent confusion but whether there is any evidence that the individual can detect them, which is required for the verbal community to be able to teach reasonably consistent self-reporting of these private events.

What are we responding to?

In sum, there are two circumstances under which we may learn to say we are confused. One involves responding to private physiological events corresponding to “confusing” situations by labeling these events as a feeling of confusion, as taught by our verbal community. The other also involves responding by labeling our reactions to certain situations as confusion, but without access to any unique physiological events. That is, it is possible that our sense of being confused does not result from actual sensations, but is merely the way we learned to label our behavior in certain contexts, just as we learned to label other aspects of our behavior under particular circumstances. (See Johnston, 2014 for additional discussion of these two alternatives.)

In other words, we may say we are confused simply because that is how we are taught to describe our behavior under certain circumstances. In fact, we often label our behavior based on not just what we are doing but on features of environmental situations at that moment. For example, we say we are grateful when someone does something nice for us or that we are disappointed when a planned event falls through. There may be a vague assumption that these labeled reactions are rooted in internal processes, whether physiological or mental, but a more parsimonious explanation is that we have learned to label our behavior in context no differently than we learn to label physical objects in the environment such as chairs or cars.

Is this a problem?

What might be the common features of environmental circumstances that prompt us to label our reactions as confusion? Skinner’s discussion of problem solving may offer some guidance (Skinner, 1953, 1969). He proposed that problems are situations in which the individual has no behavior immediately available that can resolve a condition of deprivation or aversive stimulation. For example, the discussion of problem solving in chapter 5 suggests that it is a problem when we have not eaten all day (a condition of deprivation) but have no obvious way to get food. Similarly, it a problem when we are trying to study but are distracted by a noisy party next door (an aversive condition) that we cannot shut down. The conditions of deprivation or aversive stimulation need not be obvious or substantial, however. For instance, merely hearing a familiar tune and wanting to know the artist may constitute a mildly aversive problem if you have no immediate way of answering the question. Skinner went on to define problem solving behavior as behavior that alters the situation so that the problem disappears.

From this perspective, being confused constitutes a particular kind of problem. Saying that we are confused involves situations that are aversive because a course of action that will lead to reinforcement is desired or called for but is not immediately apparent or available. For example, driving to a friend’s house without knowing the way may lead to getting lost and not knowing whether to turn right or left at an intersection, an aversive condition. Although both responses may be available, there may be no influences that signal which alternative will result in reinforcement. In this circumstance, the driver might be described as being confused. Although “confusing” situations may vary a good bit from one another in their details, they share the fact that an aversive state of affairs is created by the need to respond in the absence of influences that select responses that will result in reinforcing outcome, thereby solving the problem by terminating an aversive situation. These shared features seem to be the operational basis for our describing our reactions as being confused.

One advantage of this conception of confusion is that it does not require any assumption that such reactions are based on underlying physiological processes. The fact that we learn to label our reactions to situations with emotional referents (e.g., frustration, irritation, annoyance, exasperation, etc.) does not mean that we need to reify these terms. In other words, just because we may say that these kinds of problem situations lead to our “feeling” confused need not imply that there are actual detectable internal sensations involved, whether physical or mental.

This constraint may seem difficult to accept because our tendency to describe our reactions in terms that imply inner processes is the result of a life long verbal history. We are continually describing our reactions to daily situations with an extensive lexicon of subjective states without concern for their physical status. Again, a more parsimonious explanation is that such referents are no more that the result of labeling our behavior in context, albeit with terms that subtly suggest inner processes of some sort. The argument proposed here is that we can “feel confused” without the need to assume an underlying physiological condition or mental state. All that is required is learning to respond to “confusing” situations by describing our reaction as being or feeling confused.

But let’s not forget the role of verbal behavior

The last sentence raises the issue of the role of verbal behavior. Defining confusion as a special kind of problem in which someone faces an aversive condition defined by not being able to respond in a way that assures a reinforcing outcome would seem to apply regardless of the status of his or her verbal abilities. For example, parents who often, but not always, pick up their baby when it cries might be said to be confusing the child. When crying, the baby confronts a situation in which its history may not select a way of responding that will lead to reinforcement, in other words an aversive condition. Might these inconsistent reactions to crying be confusing the baby? Can a baby even be confused?

The environmentally based conception of confusion discussed here might suggest an affirmative answer, but we haven’t yet explored the role of a verbal repertoire, which a baby obviously hasn’t acquired. This means it has not learned a self-descriptive repertoire, which matters when considering whether someone can in any sense “feel” confused or in some way recognize a situation as confusing. Being or feeling confused in the sense discussed here involves not just a “confusing” situation but also the ability to observe and identify or label one’s reactions to the situation. Absent this skill, which may also be described as self-awareness, a baby can only react in such situations but cannot recognize them as being confusing or as causing a distinct “feeling” of confusion. To do so requires a self-observational skill not yet learned. When someone responds to a “confusing” situation with the assessment that they are confused, they are observing and labeling their behavior, even if only to themselves, which is unavoidably a verbal exercise.

Here is the test question: Can a dog be confused or show confusion? You have probably observed dogs in situations that you might judge as confusing. The dog may behave in distinctive ways to potentially “confusing” situations such as pacing back and forth, cocking its head, sniffing in a focused way, and so forth. It might be tempting for anyone not trained as a behavior analyst to describe the dog as “being confused.” (There are videos on line that show dogs reacting in this way to a magician doing slight of hand tricks that make a treat or object seem to disappear.)

As a well-trained behavior analyst, however, you would recognize the temptations of anthropomorphism and its accompanying mentalistic context. (See an earlier post titled, “The border collie and the behavior analyst” for additional discussion of this topic.) Dogs cannot feel confused any more than they can feel happy, sad, or any other apparent emotion. It may be hard to resist disagreeing with this statement because it requires you to put aside the colloquial view of canine behavior with which we’re all so comfortable. We all know that dogs behave in particular ways under certain circumstances. The problem lies in our tendency to describe or label their behavior in context in the everyday dialect we use to describe human behavior. However, no matter how obvious it seems that dogs have the same experience we do under similar circumstances, they have not acquired the repertoire to react as we do. Lacking a verbal repertoire, dogs – and babies – have not learned to observe and recognize or respond to the distinctive features of their behavior in particular environmental circumstances, a skill that requires the instructional efforts of a verbal community. It is this self-awareness that is the foundation for describing our reactions to certain situations as being or feeling confused.


Johnston, J. M. (2014). What is “__________?” August 21.

Johnston, J. M. (2016). The Border collie and the behavior analyst. October 19.

Johnston, J. M. (2018). When are feelings really feelings? November 20.

Skinner, B. F. (1953). Science and human behavior. New York: MacMillan.

Skinner, B. F. (1969). An operant analysis of problem solving. In B. F. Skinner, Contingencies of reinforcement (pp. 133-171). New York: Appleton-century-Crofts.


Are you confused?

Leave a Reply

Website Protected by Spam Master

Jim Johnston

Dr. Johnston received his doctorate from the University of Florida in 1970 and held faculty positions at Georgia State University, the University of Florida, and Auburn University. He has published both laboratory and field research with both human and non-human species on a variety of topics. He has longstanding interests in the area of developmental disabilities and founded the Master’s Program in Applied Behavior Analysis in Developmental Disabilities at Auburn University. He has served as editor of The Behavior Analyst and on the editorial boards of the Journal of the Experimental Analysis of Behavior and the Journal of Applied Behavior Analysis, among others. He has served as president of the Association for Behavior Analysis International, as well as for the Florida, Alabama, and Southeastern behavior analysis organizations, and was the first president of the Behavior Analysis Certification Board and the Association of Professional Behavior Analysts.

Are you confused? February 4, 2019