By James Macon, M.Ed., BCBA

Behavior Analysis has a saying which I’m sure most working in the field will be familiar with; “the rat is never wrong.”

It’s derived from the “skinner box,” aka an operant chamber. The idea is a researcher trying to shape certain behaviors in mice and would experiment using the skinner box to easily control variables. If the mouse failed to perform a certain behavior that the researcher was trying to shape, it wasn’t because the mouse was stupid, slow, lazy, or a “bad mouse” … rather it was the researcher’s fault for failing to systematically engineer and reinforce the behavior. Hence, the rat is never wrong.

The same phrase can be extrapolated to include people, whether it’s a young child with autism or a staff working with that child. To modernize the phrase, we’ll say the performer is never wrong. As practitioners and BCBAs, we are quick to defend our clients contextually inappropriate behavior as a function of environment and learning history, and we use science to treat those behaviors. Unfortunately, we often fail to analyze our staff’s behavior with the same rigor.

When performance problems occur or persist, staff are often blamed as being stupid, slow, lazy, or a bad employee. Not discounting that occasionally there will be bad employees, most performance issues can be resolved if the environment supports it and the correct behaviors are reinforced. Drawing from our earlier phrase, the performer is never wrong.

Organizational Behavior Management (OBM) provides some useful assessment tools to assist with this. One such tool is Aubrey Daniels’s PIC/NIC Analysis ®.   The PIC/NIC analysis assesses different influences on behavior and identifies how strong or weak the contingencies that maintain them are. Behaviors identified to have consequences that are positive, immediate, and certain (or PICs) are likely being reinforced, whereas consequences that are negative, immediate, and certain (or NIC) are likely punishing. A couple other variations include if a consequence is in the future (as opposed to immediate) and uncertain (as opposed to certain). Both of those variations decrease the likelihood that the listed consequence will be strong enough to reinforce behavior.

An example of this tool in action, suppose I have a staff who is on their cell phone when they should be working with their client. Step #1 would be to write down the problem performance, “on cell phone.” Step #2 would be to write down all of the antecedents to this behavior in the left hand column. I might identify “child on a break,” “stressed out,” “the sounds of a phone’s notification ringer,” and “no supervisor present” for starters. Step #3 I would list all possible consequences, good and bad, from the performers perspectives on the right hand side. I might identify “temporarily escape from work,” social reinforcers from Facebook, disapproval from supervisor, and performance write-up for starters. Important note, you antecedents and consequences should not be “lining up” here, so don’t worry if they aren’t.

Step #4, code whether the consequences are positive/negative, immediate/future, and certain/uncertain. For simplicity, create 3 adjacent columns to the right, one for P/N (positive/negative), another for I/F (immediate/future), and lastly for C/U (certain/uncertain). As you go down your list of consequences, continue coding how strong the contingency is. When your finished, you will notice certain consequences are PIC, which are likely to maintain that behavior. Consequences that are future or uncertain (NFU, PFU) are unlikely to control behavior.

In our example, temporary escape form work is a PIC. From the performers perspective, its positive, immediate, and certain. Receiving social reinforcers from Facebook is another PIC. Both of these behaviors are likely to reinforce being on the phone. Disapproval from supervisor is negative, future, and uncertain (NFU), and not likely to influence behavior. Same with a performance write up (NFU).

As you can see from our analysis, our staff’s behavior of “being on the phone” instead of working is occurring because of operant conditioning, and doesn’t have anything to do with them being stupid, slow, lazy, or a bad employee. For step #5, we would repeat all of the steps above for the desired behavior (working with client). Step #6, we would then re-engineer the consequences to ensure our staff is receiving PICs for the desired behavior, and not the undesired behavior (e.g., positive immediate praise from the supervisor, regular performance reviews, tangibles and verbal reinforcers for exemplary performance, etc.).

We can also use our antecedent information to help engineer performance. In our example above, having staff remove their phone or place it on silent will remove at least 1 of the antecedents. Similarly, having supervisors regularly ensure procedural integrity will also alter the likelihood of checking the phone.

The theme here of course is that rather than just relying on mentalistic labels for performance problems (e.g., the staff is lazy, unmotivated, a bad employee, etc.), we can use the principles of behavior to systematically engineer better performance. And because we are engineering more reinforcement contingencies for the desired performance, not only will the staff’s behavior improve, so will the clients. Perfect end for a picnic!