Feedback – Positive Reinforcement Revisited

The feedback post from a few weeks ago has been on my mind, as I keep looking for texts or examples that provide a bit more theory on why certain feedback models have become so pervasive, and what the reasoning behind the models is.

In this post, notes from In Search of Excellence, in which the authors come out as strong Skinner-ists (fans of the behavioural positive reinforcement theory), and a story of positive reinforcement in action.

Theory

In Search of Excellence, originally published in 1982, is one of those standard business books that all managers are supposed to read to understand how an organization becomes great. The authors researched what they identified as excellent companies and extracted eight principles that create excellence.

In the chapter Man Waiting for Motivation (it was 1982…), their expounding of Skinner is that positively reinforced behaviour will increase and increase until it starts pushing less desirable behaviour off. So if you positively reinforce enough, the behaviours you want to see will expand until they fill up all the space, and what hasn’t been reinforced gets dropped off. And, because the person has chosen by himself/herself (consciously or not) to expand one behaviour and drop off another, he/she is unlikely to go back to previous behaviour patterns (unlike behaviour changes prompted by threat, in which reversion to previous behaviour is more likely).

Why is such a straightforward guideline (positively reinforce, extensively, and don’t worry about addressing negatives) so hard to follow? I admit it’s hard to follow – I tried to give someone solely positive reinforcement after watching him run a workshop, and it was hard: feedback models have ingrained in me that I need to find things wrong and identify them.

In Search of Excellence suggests: “[Managers] either appear not to value [positive reinforcement] at all, or consider it beneath them, undignified, or not very macho” (p. 70).

Why value positive reinforcement over negative or corrective feedback?

Skinner and others take special note of the asymmetry between positive and negative reinforcement (essentially the threat of sanctions). In short, negative reinforcement will produce behavioral change, but often in strange, unpredictable, and undesirable ways. Positive reinforcement causes behavioural change too, but usually in the intended direction…

…Says Skinner, ‘The person who has been punished is not thereby simply less inclined to behave in a given way; at best, he learns how to avoid punishment.’

Positive reinforcement, on the other hand, not only shapes behaviour but also teaches and in the process enhances our own self-image…

… It nudges good things onto the agenda instead of ripping things off the agenda.

(p. 68-69)

Story

I was instructing for two weeks at the Royal Ontario Museum (ROM), running a leadership course for teenagers. The ROM’s summer program employs around 20 instructors at any given time, with about 300-400 kids coming through every two weeks. By the end of each summer, instructors who have worked all four sessions are exhausted, and the energy of instructors, volunteers, and assistants tends to wear down. But while I was working there, someone told the story of one instructor (also a leadership instructor) who started a “ROM dollars” program. Everyone tossed in $5 of actual money into a ROM dollars pot, and then the contest began. Every time you saw someone doing something great, keeping a smile on their face, still being positive and energetic and calm after eight weeks of pretty intense work, you’d give them a ROM dollar with a note of why they were receiving it. ROM dollars spread around the summer program, with people looking for behaviours that they could reinforce with a ROM dollar, and individuals looking to do things that would earn them ROM dollars. And at the end of the summer, the team that had amassed the most ROM dollars got the pot of actual money and a gift certificate at the favourite local after work hangout.

It’s a simple story, and probably many of us have been in a similar sort of reward system, but it struck me this week as the perfect story of a system that functioned solely on positive reinforcement and the participation of staff. There wasn’t a manager or higher-ups who instituted the system or who were in charge of giving the positive feedback, and there was no emphasis on corrective feedback. But an employee-initiated, employee-run system, based entirely on positive reinforcement, made for a more positive work environment for everyone when it was needed most.

Understanding Feedback

“I’ve got some feedback for you.” I’ve seen people respond to this statement with fear and apprehension, visibly tense body language, sighing… I’ve also seen people respond with excitement: their ears perk up, you’ve got their interest, and they’re eager to hear what you’re about to say.

Most of the courses I’ve taken recently have involved peer feedback between participants as we practiced our new skills. It’s amazing how much you can read about a participants’ workplace culture when you see how he/she reacts when we get to the feedback part of the day.

I decided I want to get a better understanding of what makes for good feedback. Every “how to give feedback” course/article I’ve come across gives some sort of simple formula (e.g. talk about the event, not the person, describe the situation, be specific), but little background theory on why feedback is so important, why so many people have bad associations with the word, and what actually ends up being effective.

Model 1:

The first feedback model I encountered in the workplace was behaviourist: positively reinforce behaviours you want to see repeated; ignore behaviours you don’t want to see repeated (hypothesis: ignoring the behaviour will extinguish it). Good parts of model: experientially, I agree that behaviour must be positively reinforced in order for it to be repeated and valued. Too many managers (and employees) make the mistake of ignoring positive behaviour, reasoning that if it’s not broke, it doesn’t need attention. They don’t realize that if an employee (or manager) isn’t getting feedback on what works and what’s valued, then the employee (or manager) doesn’t have the information they need to build on, and continue exhibiting, their strengths. Weak parts of model: ignoring negative behaviours doesn’t always extinguish them. How does corrective feedback fit into the behaviourist model?

Model 2:

The Sandwich model, possibly the model that I think has the most holes in it. The theory is that you start with a positive comment, so the feedback recipient feels good. This supposedly makes the recipient more receptive to the sandwich “filling” – the negative/constructive/corrective feedback. Finally, finish with another positive comment, so that you go out on a good note. Good part of model: well, at least we’ve got some incorporation of positive feedback here. Bad part of model: feedback recipients hear the opener and think “Oh no, what’s coming next…” and then they’re so busy thinking about the negative feedback in the middle that they don’t even hear the positive comment that comes at the end. Also, I think this model encourages the feedback giver to come up with positive feedback insincerely. Example stream-of-thought of the feedback giver: “I need to tell this person he’s doing x wrong. Oh, but the sandwich model! I need to first think of something I can compliment him on, and something nice to say at the end too.” Positive feedback is meaningless and insincere when it’s only being given as a way to package the negative feedback.

Sample guidelines for supportive feedback

The mediation course I took offered these guidelines for supportive feedback:

1. Look for the positives and mention them.

2. Make observations about what didn’t go well, and make suggestions for an alternative approach.

3. Describe what happened, don’t judge it.

4. Be specific.

5. Direct your suggestions for change towards behaviour the person can control.

6. End on a positive note.

Other ideas?

All of the above are sample recipes for feedback, but none are satisfactory explanations of what or why particular methodologies work; nor do they give a theoretical framework for feedback. In search of something less recipe-based and more theory-based, I’m reading Tell Me How I’m Doing (huge emphasis on supportive feedback as more important than corrective feedback) and Perfect Phrases for Building Strong Teams (I’m trying to hold off judgment until I’ve read the whole thing, but I suspect I won’t be using platitudes such as “Team feedback is a source of support and growth, not criticism” anytime soon).

More posts as (if) I turn up some better models or better ways of thinking about feedback.