Climate Week at Caravan of Dreams #8 · Cleaned Transcript
September 24, 2026 · Caravan of Dreams, East Village, NY
Climate Week at Caravan of Dreams #13 · Cleaned Transcript
September 25, 2026 · Caravan of Dreams, East Village, NY
An early-afternoon session at Caravan of Dreams, recorded on the house mic. The recording begins a few seconds into Jonathan J. Halperin's opening. Co-host Emma Orfield Johnston of Orfield Laboratories could not attend, and Steve Orfield, the lab's founder, joined by recorded video. The demonstration itself ran in the dark and in silence, so it appears here as a short description. This is a cleaned and condensed transcript: filler trimmed, crosstalk smoothed, the shape of the session kept intact, and the networking after the close left out. Names and organisations spoken aloud were checked where a source exists; where the recording was too garbled to identify one, it is described rather than named. The speakers are named; participants are not. It is not verbatim.
Download as .txtHalperin · ...community, connection and resilience, and folks have been at it here all week. This one is a little different. I'm going to talk for a couple of minutes, then we'll turn the lights way down and put you all, as participants, through a mini version of what the Multi-Sensory Climate Center can do.
I'm Jonathan Halperin. My colleague Emma Orfield, whose name you can see up there, is not going to make it. She's fine, but she was involved in a situation this morning. So it will run a little differently than we planned, and I hope you enjoy it and find it interesting. We would love your feedback.
Let me tell you about the genesis of the project and the problem we're trying to solve. We've been talking about climate for decades, and a lot of good work has been done, but the public is not fully engaged. Some people think it's a hoax, some think it's hopeless, and there are a lot of views in between. The history of changing systems at this scale requires massive public involvement, people power, and we are a very long way from that. Look at social movements: smoking, littering, civil rights. It's possible, but it takes the public getting seriously involved. So the question is, if the science is there and the media is covering it, why is the public not really engaged? We've talked with thirty-some organizations of various types that are also trying to figure out how to change that equation. Engagement, not just communications, though that's key: actual engagement.
A little background. I've done a lot of survey research over the years, as one piece of designing public engagement campaigns, including seventeen years of work in the Soviet Union and the former Soviet Union, which is an interesting place to do survey research. I executive produced a film called Hope in a Changing Climate, a story of ecosystem restoration, and part of the point was to get the agriculture people and the climate people talking to one another. We premiered it at the COP in 2009. I've done a bunch of different things.
Survey research is really valuable. I like it, and we use it. But on this topic, when people answer a survey, how do you feel about climate change, their answers are what they think they feel. That is not necessarily the same as the underlying emotions that drive behavior. Commercial marketers have understood that for a very long time. There's a reason PepsiCo doesn't put the nutrition label on the front of a bag of Fritos. They're looking for an emotional decision, and they understand how those are conjured.
So we brought in Emma and Orfield Laboratories, in Minneapolis. Almost nobody has heard of them. They have the quietest place on earth, literally, certified by Guinness and, maybe more importantly, as an accredited acoustic lab with the National Institute of Standards and Technology. They also have labs for lighting, texture, vibration, because we're multisensory people. That's what we are. We're taking the capacity they've refined over fifty-plus years to surface emotions people are not conscious of, and applying it to climate. The ultimate goal isn't a pile of data. It's analysis of data collected in a very specific way, so that organizations, thousands of them, not us, that work on climate and have constituencies can engage people in a very different and more successful way. That's the basic framework.
Halperin · I'm passing out ten sheets for each person. You don't have to fill in which session you're in; that was for an earlier demonstration where there were three sessions. Here's where this fits. In a real project we define a cohort and recruit them, and in this process you are jurors. This is a jury. We design a very specific stimulus to which we expose that specific cohort. And the form in front of you is a semantic differential: pairs of words designed to align with the cohort and the stimulus. So I'll put up some intense images, turn the lights way down, and ask you to complete one sheet per slide, marking how you feel when you see it. Then we'll come back, I'll explain a little more, and we'll hear your thoughts.
From the room · Do you need to issue a trigger warning, for anyone who might not be able to handle it?
Halperin · There's no flashing lights, none of that. They're still images.
From the room · I also feel a little bit like a lab rat. Can you help us understand what this is? It sounds fascinating, but it's a little uncomfortable.
Halperin · Let me try to ease that discomfort without getting too far into it, and we can talk more once you've seen the process. This is a demonstration. You don't put your name on anything. We're not collecting this data or analyzing it the way we would in a real project. Lab rat is a negative depiction, but in the real thing people know they're being asked to come in as members of a jury, to rate something. We know who they are, their demographics, but they don't know what they're coming in to do. You all know, so this isn't a fully legitimate demonstration in that way. Does that help? You're still in the lab, but maybe not a rat.
From the room · To put a point on it, for people who have lived through something like sexual assault or war: is there anything like that in the images?
Halperin · No. They will evoke emotions, yes. But we're not showing open heart surgery.
Halperin · Before we start: since Emma's not here, we have a short welcome from Steve Orfield.
Steve Orfield (on video) · Hello, I'm Steve Orfield, with Orfield Laboratories, the founder of the world's first and only multisensory design research lab. We've used this lab for fifty years to measure people, to measure experiences, and to correlate them, to find out what about their emotions is elicited by what they see and what they hear. What we're measuring is their unconscious feelings, not their opinions. We hope to apply this method, which we've used for thirty years, to the climate field, because we don't believe it's been used there at all, and we believe there's an extreme possibility that we'll find out new things by doing that. We hope you enjoy your experience, and that it causes you to think a bit more about a new approach to climate change research.
The lights go down. Ten climate images come up one at a time, and the room marks one sheet per image, fast, in silence. The lights come back up.
Halperin · A seemingly simple task: paper and a bunch of pen marks. We darkened the room. This is a lovely room, but it's not the way we'd set it up for a full formal project, and context matters. If you do this in a quiet, calm room, it's different from doing it in a bright room at a typical climate reception, because we are a multisensory species. So: however many people are here, times ten pages, times ten characteristics, each a bipolar word pair, which is the semantic differential, on a seven-point scale. Multiply that through and it's a lot of data. We're not going to analyze it in real time; that's not how it's done. But in a full project the capacity to learn is significant.
In a full project the images are picked very carefully. A lot of research goes into defining them: what to pick, what people have already been shown, media analysis. And the cohort is designed; this room isn't a cohort, just a random set of people who are interested in and knowledgeable about climate. Aligned with those two is the set of word pairs, not exactly opposites, that let people record a different kind of response. The ability to do that at scale and map people's emotions to climate is the core of this.
We're not trying to help organizations figure out how to talk better about climate. We've been talking about climate for a long time, and it's not going very far. The goal is to find things that are already top of mind for most people. You're not most people: you're interested in climate. Most people are not spending much time thinking about it. So we work with organizations, the thirty or so I mentioned, that have constituencies who have identified themselves: people who go to aquariums, to ocean science centers, people who go on Sierra Club outings and like trees and forests. We pick those folks as carefully designed cohorts and do this analysis, and each one is a piece of the jigsaw puzzle that is climate. Map emotion to forests, and forests to climate. Once we do that across the whole set of issues, we can understand and engage people at a different level.
The timing of how I asked you to do this is calibrated too, and since Emma's not here I won't go too far into it. The idea is that you record something before your brain intervenes, the emotion that hits you first. The brain's interpretation is an opinion, a thought. We want the deeper, more visceral emotion. One of Emma's examples: you're at a movie and something makes your eyes well up. That's an emotional response. Then you walk out and your friend asks how you felt about it, and your brain explains. Those are related, but they are not the same thing.
The Orfield folks call the methodology perceptual market research, and they invented it. Each step builds out in a lot of detail, and depending on the project, each can take months. From my perspective the key piece is the back end: how we get what we learn to the organizations that can use it. We know the methodology works. It has worked for Harley-Davidson, for Herman Miller, for 3M. They've done fifty years of research, mostly for corporations, and we're taking it into the climate realm.
Halperin · So you've been through a shortened, miniature version of what the Multi-Sensory Climate Center is built to do. Questions, thoughts, ideas? Did you feel like a lab rat?
From the room · Even with the lights down, I felt like I went through a filter, translating how I felt into where I am on the scale. Has this been compared with biometric data, like my heart beating faster? I know a lot of that can be the same physical reaction, interpreted differently by the brain depending on context.
Halperin · Good question. I don't know for sure. Steve is a scientist's scientist, with lots of published and peer-reviewed work, and the papers are linked in the deck. I believe it's been checked, or big companies wouldn't keep going there. Another example: among the thirty-ish groups we've been talking with is an institute that studies consciousness, with a neuroscience lab that studies chills, the kind that lasts a fraction of a second and takes you out of your present reality. Is that physical, biological, neurological? They've worked out how to extend that moment. The researcher I spoke with reached behind him to a shelf and pulled out a device the size of a deck of cards that goes on someone's back, calibrated to where the charge lands, to lengthen the chill. There are big applications for people with trauma. Humor is somewhat similar: you're going along and then there's a turn. How do you extend that turn? So there's a lot of research we're not commissioning but are bringing people in from, and we're building an advisory group, from forests to psychology.
From the room · What have you discovered so far?
Halperin · That it's a big and complicated project that is not fully funded. We have partners for pilots and other potential projects, but we're not running full sessions yet.
From the room · Have you found good practices or visuals to start combating compassion fatigue, whether for humanitarian crises or other issues? I feel like that's a huge problem.
Halperin · The simple answer is no. A more thoughtful answer: we could put together a cohort of people who have been through a set of experiences, like helpers who go through trauma over and over, and be really careful about the order and the images we show. But we don't have any guidance on that at this point.
From the room · A reflection on the exercise. On some scales I had a clear, instant gut response. On others I had to sit and think, which is more like activating the rational brain. With some images it went too fast: I hadn't figured out how I felt yet. I wonder how those two kinds of decision can be reflected.
Halperin · If you'd mark the ones that were tricky, the ones you had to think about, that would be great feedback.
From the room · Mostly it felt cumulative. The negative images accumulated in my body, so the later images probably had more of that. Even the positive ones became overwhelming, and it made for anxiety and detachment.
Halperin · Interesting. I appreciate you sharing that.
From the room · So how do you make sure the data for each image isn't biased because it comes after the others? It's like the order of questions in survey research.
Halperin · Theoretically, an interesting question, and I don't know the answer. It's really useful. It may be the kind of thing where you run multiple cohorts, or give people the images in different orders.
From the room · I don't know if this is useful, but twenty to thirty percent of the population has heightened sensitivity, so you might look at cohorts according to that.
From the room · There's research on it: differential susceptibility, the highly sensitive person research by Elaine Aron, and something called the orchid hypothesis.
From the room · Could you share an example of how the research leads to a use case?
Halperin · Sure. Orfield has not done issues like this, but here's a true case. Harley-Davidson comes to Orfield Laboratories and says: we've got a big problem. EU regulations require us to lower the decibel level, the loudness of the bike. We can do it, but we'll lose huge market share in the US, because people buy the bike because it's loud. What do we do? Orfield asks: how do you know they like loud? Have you done research? They didn't really have an answer. So they interview a whole host of Harley riders, take riders out to a racetrack, and monitor all sorts of aspects of the experience and the engine. Long story short: every rider will tell you what they like is that it's loud, along with the price, the shape and so on. But what Orfield told Harley is, you're fine, because what the riders think they feel is loudness, and it's not. It's low-frequency vibration. It isn't the decibel level. So again: what's the stimulus, what's the cohort, and how do you identify the underlying, preverbal, not-thought-about experience?
From the room · So how would you do that with climate?
Halperin · We don't have a case, because we haven't done a full study. But that it's doable is clear. The first thing we'll do is pilots, to refine exactly how to make it work on an issue, and an issue is not tangible. Methane is a great issue, but most people have no experience of methane. That's one reason we're doing trees: most people everywhere have an experience with a tree of some kind. Methane and trees sit at opposite ends of that spectrum, and both are really important for the planet.
From the room · I used to work at a residential electrification nonprofit, and a huge problem was adoption of technologies that are already cheaper and already better. That's a clear extension of this research. Why don't you have a heat pump? The deeper question you're pointing at is identity change: how people develop their relationship with the environment and with each other.
From the room · I understand change is what we're trying to do. My challenge is: what are we doing here? How does the data we're getting here get to change of identity, change of needs, change of roles, the chaos and uncertainty we experience? I'm trying to understand the connection.
Halperin · You don't get anything from the data analysis of this; it's a demo. But for organizations working on climate, we're offering to build emotional profiles of their constituencies. It's ambitious, beyond what we can do right now, but think of a Gallup on emotion, not opinion. Then we show organizations how to use the profiles to engage whatever constituency they work with more effectively. It is marketing, but it's not just marketing or messaging. It's the whole strategy for engaging a population, because we understand at a completely different level why people go to the science center or the natural history museum, and why they care or don't.
From the room · And this is the methodology they would use to capture that?
Halperin · Yes. This was a demo, but it's what the laboratory has been doing for years.
From the room · Like the first question you got: if I sensed at all that an image was AI-generated, I immediately didn't trust it and disengaged from the process. There were at least three. The baby, and the coral one in the sea. Some were more obviously AI than others.
From the room · I agree. Once you think, wait, that's not real, you wonder what you're doing, and then I question all of them.
Halperin · I agree with you. When we do real projects, the amount of time, effort and money that goes into making sure those three things line up, the stimulus, the cohort and the language, is large. The stimulus here was visual, but it could be a different kind. And these were not up to that.
From the room · Today you used a paper form. Does the center also use other methods to get that data?
Halperin · I'm not an Orfield person, but I believe the answer is yes. On instrumentation, it's quite a place, and they do a lot of work with veterans with PTSD. One example of their capacity: medical instruments with a slight vibration that surgeons can't quite detect but know is some sort of issue. They have what they call a sound telescope, with many channels, that can measure it. Although, the first time I went to the lab, in Minneapolis: a cinder block building, boring, ugly. You go into the lobby and there are two blown-up record covers, five feet across. One is Bob Dylan's Blood on the Tracks and the other is a Cat Stevens album. I thought, this is a scientific laboratory? They're there because the building was a recording studio, and both of those albums were recorded there. It may be the only lab in the world that can do this kind of multisensory analysis.
From the room · The testing is done with a kind of sensory deprivation. But for someone taking this in day to day, their senses aren't deprived. How do you reconcile that, and what's the difference in data and experience?
Halperin · We're not trying to replicate the experience people have outside. We're specifically taking them into a more isolated situation that limits the number of variables driving their emotions: smelly, hot, cold. We try to neutralize every variable we can. And I cannot fully explain why, since I'm not the scientist, but the magic number is thirty-two. Every jury they've assembled for the last thirty-five years or so is thirty-two people. From that, they're able to take subjective data and turn it into quantitative data. I'm sure some projects have failed, but the big corporations keep going back to them.
From the room · This is fascinating, and I can kind of surmise why you got interested, but I'd love to hear from you: why are you doing this, and why is the lab doing it, especially if no one is funding it yet?
Halperin · Good question. The lab is doing it because I went out there, we spent time together, and they want to have an impact on the world beyond the corporate clients. They believe, for all the reasons we're interested in climate, that it's an existential and multigenerational crisis. For me, I'm drawn to complexity: US-Soviet relations, ecosystem restoration. I like trying to understand how systems work and where we can change them, and helping other people understand that. I've done a lot of work on democracy, which I believe in. But none of that will have traction, whatever we work on, because climate is going to affect absolutely everything, and already is. I've got kids, and friends with kids, and what we pass on to them is a big driver for a lot of people.
But I got to this because I was frustrated with the survey research. George Mason and Yale and Potential Energy do tremendous research, and the public tells them what they think. There are lots of organizations working really well, but why is it not grabbing people? With civil rights we got to a point where it was not okay anymore: making Black people sit at the back of the bus. People said no, fix it, that's not something we can stand for anymore. We need to get to that.
From the room · What was it about that situation, if you're comparing the two?
Halperin · None of the comparisons are totally aligned: place, time, politics, policy, power dynamics. I don't know all of that research, but I know some of it, from people who study the sociology of systems change, cultural evolution and individual behavior. There's a lot of literature on what ingredients have to come together, and in what sequence, and it's different for littering and civil rights. But both changed. I'm not expert in that literature, but we've started consulting it, and as the project fills out we're building a set of key advisors who know it cold.
From the room · I'm curious why anger was not in the list of emotions, because for a lot of these images I felt it, particularly the one with the child. As a side note, I think there are ethical issues in showing some of these.
Halperin · I think the answer is that the Orfield team doesn't know climate that well yet. My team wasn't as involved as it normally would be, and we don't have all the advisors in place yet to vet the alignment of the stimulus, the cohort and the language. It's a really tricky question: ten items, seven points, getting it down to twenty words. That's not an easy task, because climate is so big. That's why we're doing very specific pilots before we go national.
From the room · I'm curious whether anger is a useful emotion. This is a company that was made to help companies sell products, right?
From the room · Measuring it may be relevant even if it doesn't motivate you. It might just be paralyzing.
From the room · Outrage is one of the strongest emotions tested for motivating people to action. Disgust is not. Disgust is paralyzing; outrage is very useful. That's why social media is so good at promoting outrage.
From the room · I would put anger in as a primary emotion, like sadness and fear, as opposed to the more complex emotions on the sheet.
Halperin · It's a bit of a Russian nesting doll: which is inside which. The psychology community isn't settled on it, even the specialists. But it goes back to the purpose. For example, Orfield's reputation with architects is terrible. Architects hate the Orfield folks, because when a company's senior leadership calls Orfield in to check a new building or a renovation, Orfield tests whether the people in it will be comfortable: is the light right, and so on. Architects tend to spec things without emotional data on the comfort they provide. And almost inevitably, I've heard them say, they can reduce the budget that was planned and get better satisfaction from the people who work there.
Halperin · I'm loving this, but it would be great to give people a chance to mingle and connect. I'm passing around a short summary of what we've done, with my email and Emma's, so if you want to connect, collaborate, or ask anything else, push on me. Thank you.