Use Reference Incidents to Clarify Communication of Experience

5Aug
2016

A common source of confusion in conversation is when the participants aren’t talking about the same thing. If the thing in question is a concept, it can be helpful to taboo your words so that you have to unpack the concept. Does a tree falling in an empty forest make a sound? Well, depends on what you mean by “sound”.

In other situations, where the conversation is about communicating our experiences of each other (aka “feedback”) there’s another habit that I find useful, that I call using reference incidents. The idea is to find or create a shared example of a particular behaviour or experience in question, with features known to both parties, so that you can compare notes on the incident and reveal your perspective to the other person in a way that they’re able to integrate it. This is related to the virtue of narrowness: “What is true of one apple may not be true of another apple; thus more can be said about a single apple than about all the apples in the world.”

A short example from my life of how this sort of confusion shows up. I was talking with a housemate, and she said to me, “This experience right now reminds me of this time with Jean,” and went on to describe that incident. Then, some minutes later she mused, “…yeah, when I spoke I think I was feeling discomfort.” At this point it would be easy to assume that she meant that she was feeling discomfort in talking with me just then, but she also could be still thinking of the conversation with Jean. (Or I could have made the opposite assumption).

Of course, in this case it may simply be both. But I asked about it, because in making sense of her talking about her discomfort, I’m bringing a lot of background context based on my own understanding of the conversation in question, and so if she and I are thinking of different conversations, then it’s going to be hard for me to make accurate inferences about what she’s saying. This makes communication hard.

Here’s a more extensive, made-up example:

Jack is talking with his housemate, Saara, who lives in the room next to his. She remarks to him that she’s sensitive to noise, and says that Jack is kind of loud.
Jack is kind of annoyed, because she’s told him before that she goes to bed early and doesn’t like when he plays music after that point, and so he’s stopped doing it, and has also tried to be aware of when she might be napping midday.
Jack therefore protests: “I’ve actually been trying to be quieter.”
…but Saara was actually talking about finding his voice loud when he’s talking to people in the kitchen.

This is a pretty extreme misunderstanding, so if they talk much more about it at all, they’re likely to realize that. But if they don’t, then a few things might happen:

Jack might get frustrated at Saara for not recognizing his efforts to date.
Saara might feel shut down in her attempt to give feedback, and confused because it doesn’t seem like Jack has been lowering his voice.
Jack might feel fine but not actually learn anything from what was intended to be new information.
Saara might feel like Jack doesn’t have anything to learn from what she has to say, when in fact he does.

Kind of a mess.

Incidentally, this is also a case where one of our favorite growth-mindset reframings can be helpful: speaking of things in the past tense. In the example above, Saara says “You’re kind of loud.” If instead, she said “You’ve been kind of loud,” then it suggests more clearly that there are specific experiences of loudness that Saara is referring to, and invites the question: “when?”

Since this is a fictional situation, we can have introduce causal paradoxes without issue. So let’s say that Jack has read this very blog post, and so he does ask that question.

Jack says, “Wait, can you give an example of a time I was loud? I want to know what was going on in my head at the time.”
Saara says, “For instance, earlier this morning you and Kelsey were making breakfast and you were talking really loudly. I was sitting at the table and found it almost painful, and definitely very distracting.”
The first thing Jack says is of course, “Ohhhh, loud as in voice. I thought you meant when I’m playing music in my room. Well good thing I asked.”
Saara says, “Huh, good thing!”
Jack then ponders, “So yeah, what was going on in my head at that moment? Well, I was definitely excited, and mostly paying attention to Kelsey. I think I knew you were in the room, but just barely, and it didn’t occur to me to think of you as someone being affected by my talking. I guess maybe I want to try being more aware of both the volume of my voice but also the presence of people in a room who aren’t part of the conversation.”

Because Saara has named a specific reference incident, Jack is able to access an enormous wealth of information that wouldn’t have otherwise been available. Saara didn’t know that the update Jack needed to make was to become more aware of who’s in the room. She might be able to guess, of course, if she had multiple experiences that seem that they could all be explained by that model.

But otherwise, what Saara had to offer was to point at that specific incident and say “this is what it was like for me (loud, painful, distracting)”. It took Jack’s access to his own experience of the reference incident for him to realize that.

So the general principle here is that if you’re giving interpersonal feedback, it’s exceedingly valuable to have a specific shared incident to point at.

Brienne Yudkowsky has a couple of videos on the subject of Cognitive Trigger Action Plans, aka CTAPs. Here’s the first one. The idea is to have a little if-then trigger and action, where the “if” is a particular kind of feeling or thought. Specific feelings and thoughts can be hard to notice, so a large part of the technique is just learning to notice them.

Similarly, if you’re giving interpersonal feedback, it’s way easier for people to figure out what algorithms their minds are running and how to tweak them if they have a shared reference incident in mind. Ideally, it’s a specific, recent incident, but an older one can work, and sometimes you might not be able to pinpoint a single instance.

Having a shared reference incident helps you integrate what someone else is saying into a coherent understanding. Part of what that entails when it comes to making sense of someone’s experience of you is to link it to your own experiences of similar situations.

So that’s the concept.

Use reference incidents to make interpersonal feedback more clear. Share your experience of the situation, and invite the other person to explore what their experience was. Importantly, don’t assume that you know why they did what they did.

You can stop reading now, or read one final example (actually a meta-example) from a conversation I had a few months ago with my friend Shane (the guy who came up with the “shoals” concept).

We were at the Focus Week event that I co-hosted, and a few of us had gotten talking during an evening break about the prospects of a video chat app, and the importance of tone of voice and body language. I was saying that these were overrated, in part because I’d recently read this article about how with the current technology, texting can feel more intimate than talking on the phone, given calls dropping and so on.

Anyway, someone cited that classic figure: “93% of communication is tone of voice and body language”. I don’t even know what that claim is supposed to imply, and so my response to that being said was to throw my arms over my head and exclaim “What the fuck?!”

People of course then asked me about my take on it, at which point I was able to use my very act from a moment ago as an example that we could all talk about. I said, “Imagine that I hadn’t said anything, but had just waved my arms in the air. Would you have gotten half of my meaning?” They weren’t sure. Then I said “What if I’d waved my arms in the air and exclaimed ‘oh, then what are we waiting for?!’ with the same tone?”

The following day, I was having an unrelated conversation with Shane and I wanted to tell him about the concept of reference incidents. He had gotten the basic jist, but I figured it would be helpful if I could point at a specific reference incident that he had experienced.

So I referred him to yesterday’s conversation and the “What the fuck?!” exclamation I’d made, and pointed out how it had served as a constructed reference example:

If we’d tried to debate the relevance of tone of voice and body language in abstract, we might have had totally different ideas in our head of what sort of situation we were talking about. Since we were instead able to use a very specific example that was present in each of our immediate memories, we could try out models to explain it and be confident that we were talking about the same thing.

Cool, eh?

I hadn’t been thinking that far in advance when I made the “What the fuck?!” remark, but it does feel sort of like a thing I did on purpose, in the sense that I try to generate salient experiences in general.

I posit that if you think that reference experiences are cool, you will also think that the sparkly pink ball thing is cool. It’s also about overcoming illusion of transparency by forcing specificity.

And being able to cite a shared experience is totally an example of the sparkly pink ball thing—you’re saying “do you remember that time when X?” and the person usually either has a relatively rich memory they’re able to bring up, or essentially nothing. So if they do remember, you can talk about it with a fair bit of assumed shared understanding.