WEBVTT

NOTE Machine-generated transcript; not human-reviewed.
NOTE Canonical transcript: https://opentheory.net/transcripts/tanha-with-neuroscientist-michael-johnson/

1
00:00:00.000 --> 00:00:08.500
Hello and welcome to Imagine an Apple, the podcast about our inner mental worlds. My name is Francis and today our guest is Mike Johnson.

2
00:00:08.900 --> 00:00:09.340
Hi everyone.

3
00:00:09.500 --> 00:00:12.380
Also with me is Vin, my co-host. Hello Vin.

4
00:00:12.980 --> 00:00:21.880
Hey Francis. Hi Mike. So Mike is, uh, formerly worked at the Qualia Institute but now is currently doing other stuff.

5
00:00:21.880 --> 00:00:29.320
But today we will be interviewing Mike about his internal experiences, specifically about the stuff that he specializes in.

6
00:00:29.360 --> 00:00:34.140
Tanha, qualia, and active influence among some of the things we'll be talking about.

7
00:00:42.660 --> 00:00:44.060
So

8
00:00:44.560 --> 00:00:51.800
So what we're interested in talking about today is inner experiences and how you've

9
00:00:51.880 --> 00:01:00.860
might experience some of these, um, phenomenal, uh, your phenomenology, whether it's more visual or auditory or, or, um, olfactory.

10
00:01:01.060 --> 00:01:07.840
And I was wondering if you have, um, special kinds of phenomenology that you think are uncommon in the general population.

11
00:01:08.060 --> 00:01:11.300
Hmm. Thanks. That's a, that's a really good question.

12
00:01:11.580 --> 00:01:20.320
Um, so I would open with, uh, you know, no one exactly knows how common or uncommon their phenomenology is.

13
00:01:20.440 --> 00:01:24.000
but we can sort of analyze base

14
00:01:24.000 --> 00:01:33.840
base rates and different sort of physiologies. And yeah, I was actually talking about this with

15
00:01:33.840 --> 00:01:40.800
a friend today. And I was saying, well, I think that I have a somewhat unusual

16
00:01:42.940 --> 00:01:49.120
phenomenology. And they were saying, okay, cool, Mike, why do you think that? What makes you say

17
00:01:49.120 --> 00:01:59.660
that? And I actually wrote a decently long article about something very similar. It's called

18
00:01:59.660 --> 00:02:07.720
autism as a disorder of dimensionality. And it basically makes the case that people vary in a

19
00:02:07.720 --> 00:02:15.300
very interesting way. So part of physiology is understanding dimensions of natural variation.

20
00:02:16.440 --> 00:02:18.880
And kind of the interest,

21
00:02:19.120 --> 00:02:25.620
give rise to interesting personality types, or in this sense, interesting phenomenologies.

22
00:02:25.780 --> 00:02:37.060
And my thesis in that piece was that people vary in terms of literally how many neurons they have

23
00:02:37.060 --> 00:02:46.900
in a unit volume of brain mass. And this is based on some research, some autopsies of

24
00:02:47.680 --> 00:02:48.900
children with autism.

25
00:02:49.120 --> 00:02:58.800
And basically, that, and this is actually a pretty crazy, crazy result. They did autopsies

26
00:02:58.800 --> 00:03:08.200
and found that autistic children had on average, 67 more, 67% more neurons in the same volume of

27
00:03:08.200 --> 00:03:16.300
prefrontal cortex, which is a lot. It's not just 1% or 10%. It's 67% more.

28
00:03:16.480 --> 00:03:19.040
Um, so that got me,

29
00:03:19.120 --> 00:03:26.020
I'm sort of thinking about, okay, well, what does that mean? And can we sort of derive

30
00:03:26.720 --> 00:03:32.720
interesting things from this sort of one factor model? And my thesis was that, yes, we can.

31
00:03:33.620 --> 00:03:41.840
And I guess I'd sort of frame that as, if you, if you take a, so everyone's kind of familiar with

32
00:03:41.840 --> 00:03:48.140
how LLMs work, at least a little bit these days. And, and sort of the, the challenges of,

33
00:03:48.200 --> 00:03:49.100
if you sort of, if you sort of, if you sort of, if you sort of, if you sort of, if you sort of,

34
00:03:49.120 --> 00:03:54.460
sort of have a certain number of parameters in your LLM, then you have certain alignment

35
00:03:54.460 --> 00:04:01.500
challenges. Certain things work to align the LLM and certain things maybe don't work as well.

36
00:04:01.700 --> 00:04:11.940
And in this case, basically you'd be jacking the parameter count by 67%. And so you get into

37
00:04:11.940 --> 00:04:19.860
a different kind of, different kinds of failure mode. And I promise this is answering your

38
00:04:19.860 --> 00:04:31.560
question. Yeah. So the observation that I made in the piece was that the natural sort of circuitry

39
00:04:31.560 --> 00:04:41.160
that evolution gives us sort of has pre-built optimization into it. That this is a

40
00:04:41.940 --> 00:04:51.380
very finely tuned evolved circuit that produces a specific result. But basically if you make

41
00:04:51.380 --> 00:04:56.940
these networks thicker, if you take an optimized circuit, I don't know if you're familiar with

42
00:04:56.940 --> 00:04:59.960
in computing, there are ASICs,

43
00:05:00.000 --> 00:05:11.260
application-specific integrated circuits, basically a custom chip for a custom task. And it does one thing, but it does this one thing really well, so like a Bitcoin miner.

44
00:05:11.900 --> 00:05:23.240
And then you have FPGAs, field programmable gate arrays, that basically, it can be anything. This is a chip that you can reprogram at the hardware level.

45
00:05:23.240 --> 00:05:31.260
You can make it look like any other chip, or you can sort of solve any problem with hardware, but you actually have to do the programming.

46
00:05:31.460 --> 00:05:45.680
And so, if this sort of autists have more thicker networks, we can say, more neurons and more connections between neurons,

47
00:05:45.820 --> 00:05:49.840
then their circuits are more like an FPGA than an ASIC.

48
00:05:49.940 --> 00:05:53.220
More like a general unoptimized circuit.

49
00:05:53.240 --> 00:05:59.040
More like a generalized network than sort of a sleek, highly tuned, thin network.

50
00:05:59.760 --> 00:06:00.280
Right.

51
00:06:00.360 --> 00:06:01.300
So, yeah.

52
00:06:01.700 --> 00:06:05.420
So, in this, so I'll pause a little bit. Does that all make sense?

53
00:06:06.100 --> 00:06:08.260
So, I'm a little bit lost at the moment.

54
00:06:08.380 --> 00:06:13.000
I'm not entirely sure what an ASIC is, or the other thing that you mentioned, an FPG.

55
00:06:13.840 --> 00:06:21.480
And not entirely sure how emotion, how this relates to emotions and qualia.

56
00:06:23.240 --> 00:06:25.980
So, expand a little bit more what you mean by like dimensionality.

57
00:06:26.300 --> 00:06:30.980
What does that mean for autistic brains to have different dimensionality?

58
00:06:31.120 --> 00:06:32.220
Yeah, yeah, you bet.

59
00:06:33.400 --> 00:06:38.160
So, the, so an ASIC is a specialized chip.

60
00:06:38.620 --> 00:06:40.000
It's very optimized.

61
00:06:40.300 --> 00:06:43.780
And an FPGA is a much more generalized chip.

62
00:06:43.920 --> 00:06:44.940
It's not optimized.

63
00:06:45.180 --> 00:06:46.780
But you can program it yourself.

64
00:06:46.980 --> 00:06:52.580
And the sense is that if a brain has sort of extra neurons in it,

65
00:06:53.240 --> 00:06:56.120
it's more like an FPGA than an ASIC.

66
00:06:56.260 --> 00:06:58.680
It has a lot of potential in it.

67
00:06:58.820 --> 00:07:01.640
But it's not particularly optimized.

68
00:07:02.040 --> 00:07:02.300
Right.

69
00:07:02.740 --> 00:07:04.260
But what is it optimized to do?

70
00:07:04.500 --> 00:07:05.320
Yeah, yeah.

71
00:07:05.480 --> 00:07:10.760
So, we generally, like a lot of circuits are pretty optimized in the brain.

72
00:07:10.960 --> 00:07:13.400
Like we have circuits to detect faces.

73
00:07:13.660 --> 00:07:16.240
We have circuits to detect noises and scary noises.

74
00:07:16.580 --> 00:07:21.180
We have circuits to figure out which foods are tasty and which are not.

75
00:07:21.980 --> 00:07:23.220
And the, the ASIC is a very well-known tool.

76
00:07:24.240 --> 00:07:26.820
people on the spectrum, on the autism spectrum,

77
00:07:27.480 --> 00:07:34.760
might have to do a little bit of post-production tuning on their circuits.

78
00:07:34.820 --> 00:07:41.640
They're not so optimized because it's just they have more neurons

79
00:07:41.640 --> 00:07:43.720
than what evolution was planning for.

80
00:07:44.040 --> 00:07:44.840
Right. Okay.

81
00:07:44.920 --> 00:07:48.880
So does this translate to a different inner experience, so to speak?

82
00:07:48.960 --> 00:07:52.940
Do they have qualia that might be different from someone who's not autistic?

83
00:07:54.280 --> 00:07:56.320
You know, I want to say yes.

84
00:07:57.100 --> 00:08:04.460
And I think that part of that is every sort of every ASIC is going to be

85
00:08:04.460 --> 00:08:08.680
every sort of highly evolved circuit is going to be very similar.

86
00:08:08.860 --> 00:08:14.240
But every FPGA is going to be wired up a little bit differently.

87
00:08:15.220 --> 00:08:21.760
And so you're going to get a lot more diversity in sort of the solutions

88
00:08:21.760 --> 00:08:23.480
that the brain finds.

89
00:08:24.240 --> 00:08:25.140
Oh, okay.

90
00:08:25.920 --> 00:08:32.020
So I guess this would like, I don't particularly make autism a part of my identity

91
00:08:32.020 --> 00:08:33.480
or not part of my identity.

92
00:08:33.580 --> 00:08:34.540
It's just the word.

93
00:08:34.800 --> 00:08:42.480
But I do think that like this, I sometimes interact with the community

94
00:08:42.480 --> 00:08:45.920
and there's this focus on neurodivergence.

95
00:08:46.240 --> 00:08:53.500
And I think that's interesting because it's like everyone with autism is going to be,

96
00:08:53.500 --> 00:08:54.220
you know,

97
00:08:54.240 --> 00:08:56.240
different than people without autism,

98
00:08:56.400 --> 00:08:59.120
but also different from other people with autism.

99
00:08:59.260 --> 00:09:03.560
It's like if you have a very small circuit,

100
00:09:03.660 --> 00:09:04.980
there's one way to make it work.

101
00:09:05.120 --> 00:09:06.960
If you have a very big circuit,

102
00:09:07.120 --> 00:09:09.580
maybe there are a hundred different ways to make it work.

103
00:09:09.660 --> 00:09:11.680
And maybe you found solution 57,

104
00:09:11.880 --> 00:09:14.160
whereas someone else found solution 62.

105
00:09:15.220 --> 00:09:21.360
And I think that that may sort of show up in sort of how sensory experiences

106
00:09:21.360 --> 00:09:23.780
present to people.

107
00:09:24.240 --> 00:09:24.500
Right.

108
00:09:24.620 --> 00:09:27.780
So here you're talking about the phenomenology of direct sense.

109
00:09:27.920 --> 00:09:31.920
So how people, when they're actually seeing something, experiencing it,

110
00:09:32.040 --> 00:09:34.200
is that what you mean would vary?

111
00:09:34.740 --> 00:09:35.200
Yeah.

112
00:09:35.700 --> 00:09:39.560
I think that you actually get different results in phenomenology.

113
00:09:39.760 --> 00:09:40.320
Okay.

114
00:09:40.400 --> 00:09:45.740
And in terms of the kind of more imaginative, internal phenomenology,

115
00:09:45.840 --> 00:09:49.980
so like inner imagination and inner voices, that kind of thing,

116
00:09:50.160 --> 00:09:52.700
do you think use of those would vary?

117
00:09:52.840 --> 00:09:54.220
Or is that like a separate question?

118
00:09:55.640 --> 00:09:58.160
I think it's all linked.

119
00:09:58.340 --> 00:10:00.000
So yeah, I would also

120
00:10:00.080 --> 00:10:02.300
apply this to that frame as well.

121
00:10:02.440 --> 00:10:18.860
So imagination often has a practice component as well. And for many people, they engage in some kind of attentional training. Meditation being a pretty good example of this, but other techniques, alexander techniques, various other things also come into play.

122
00:10:19.120 --> 00:10:22.940
I'm aware that you are a meditator. Am I correct on this?

123
00:10:23.200 --> 00:10:23.880
Yeah, that's right.

124
00:10:23.880 --> 00:10:35.060
Do you think that your meditation practices lead to different inner experiences, whether it regards to like your imagination or how you experience qualia differently?

125
00:10:35.580 --> 00:10:46.380
Yeah, I would say so. If I had to say some words on that, I do think that just practice looking at things makes them more clear.

126
00:10:46.580 --> 00:10:51.580
And I would also say that just like you mentioned alexander technique.

127
00:10:51.660 --> 00:10:53.360
And I guess,

128
00:10:53.880 --> 00:11:01.360
especially lately after my latest piece on vessel computation, I've been sort of practicing this.

129
00:11:02.320 --> 00:11:05.660
There's this phrase, opening the hand of thought.

130
00:11:06.020 --> 00:11:11.860
And I think that practicing that is a very powerful move.

131
00:11:12.320 --> 00:11:21.620
And often there are clenches and clamps in our phenomenology and just sort of practicing, okay, like, what would it be so bad?

132
00:11:22.260 --> 00:11:23.600
Would anything bad happen?

133
00:11:23.880 --> 00:11:28.160
Would it be so bad if I just sort of released and just tried to be an open experience?

134
00:11:28.560 --> 00:11:29.980
And I think that's very powerful.

135
00:11:30.340 --> 00:11:31.860
Yeah. Opening the hand of thought.

136
00:11:31.920 --> 00:11:32.780
I really like that phrase.

137
00:11:33.300 --> 00:11:37.340
You mentioned vasocomputation, which is something that I've seen you written about.

138
00:11:37.520 --> 00:11:44.400
Can you tell us more about what vasocomputation is and how that relates to TANA, which you mentioned quite a lot in your blog post?

139
00:11:44.660 --> 00:11:45.880
Sure. Yeah.

140
00:11:46.100 --> 00:11:51.660
Well, I'm a bit afraid that I'm kind of throwing listeners into the deep end here.

141
00:11:53.200 --> 00:11:55.380
That's okay. We'll unpack it once we're done.

142
00:11:55.660 --> 00:11:56.980
Nice. Nice. Good.

143
00:11:58.040 --> 00:12:02.760
So I wrote this piece called Principles of Vasocomputation.

144
00:12:02.860 --> 00:12:07.680
And it's basically a unification of three or four domains.

145
00:12:07.940 --> 00:12:11.580
And the first domain is Buddhist phenomenology.

146
00:12:11.680 --> 00:12:16.160
And especially this term that the Buddhists call TANHA.

147
00:12:16.280 --> 00:12:21.060
And this is often translated as grasping or thirst.

148
00:12:21.660 --> 00:12:23.460
Or clenching or desire.

149
00:12:23.780 --> 00:12:33.080
And according to the Buddhists, it's responsible for maybe 90% of all sort of moment by moment.

150
00:12:33.340 --> 00:12:35.820
And so it's kind of, it's a big deal.

151
00:12:35.920 --> 00:12:36.380
It's important.

152
00:12:36.760 --> 00:12:39.900
Just to make it clear what you mean by TANHA.

153
00:12:40.060 --> 00:12:44.800
When you experience it yourself, can you describe how that comes to you?

154
00:12:45.000 --> 00:12:49.900
Do you get like, is it like a verbal thought or do you see things?

155
00:12:51.660 --> 00:12:52.920
Or is it just the experience of having TANHA?

156
00:12:53.040 --> 00:12:58.960
Yeah. So there's, I would call it a contraction in awareness.

157
00:12:59.220 --> 00:13:06.500
And there's both the general contraction and a contraction around a specific part of awareness.

158
00:13:06.760 --> 00:13:13.700
And you can kind of gauge how flat your phenomenology is in a way.

159
00:13:13.800 --> 00:13:18.000
And when TANHA is going,

160
00:13:19.720 --> 00:13:24.940
your phenomenology feels more lumpy, I guess I would say, for lack of a better word.

161
00:13:25.120 --> 00:13:25.880
Cool. Okay.

162
00:13:26.360 --> 00:13:33.180
So how does TANHA relate to this thing called free energy and active inference?

163
00:13:33.380 --> 00:13:36.220
And can you help unpack those terms and what they mean?

164
00:13:37.000 --> 00:13:37.940
Sure, sure.

165
00:13:38.300 --> 00:13:43.740
So principles of vessel computation connects three domains.

166
00:13:43.920 --> 00:13:45.940
And one is Buddhist phenomenology.

167
00:13:46.080 --> 00:13:47.980
And the second domain is TANHA.

168
00:13:48.000 --> 00:13:51.100
And the third domain is the active inference framework.

169
00:13:51.710 --> 00:13:58.480
Or the more full mouthful there would be the free energy principle active inference framework.

170
00:13:58.620 --> 00:14:03.620
And so this was developed by Fristen and others.

171
00:14:03.800 --> 00:14:09.240
And it's basically a framework how the brain is a prediction machine.

172
00:14:09.400 --> 00:14:11.280
And we sort of hallucinate our reality.

173
00:14:11.420 --> 00:14:13.600
We're always trying to predict our sensations.

174
00:14:13.960 --> 00:14:17.700
And we live in that story, we can say.

175
00:14:19.800 --> 00:14:25.140
And active inference is another layer on top of this that says,

176
00:14:25.240 --> 00:14:28.240
not only do we make these predictions all the time,

177
00:14:28.300 --> 00:14:30.660
not only do we try to predict our sensations,

178
00:14:32.040 --> 00:14:41.720
but sometimes we predict false sensations and we hold these predictions until we act in the world to make them true.

179
00:14:42.120 --> 00:14:45.720
So this is a very subtle, important thing, I think.

180
00:14:45.840 --> 00:14:47.980
That, for example,

181
00:14:48.000 --> 00:14:54.720
you know, if I'm thirsty, maybe I'll predict, oh, I'm not thirsty.

182
00:14:54.920 --> 00:14:58.280
I have the taste of water in my mouth and down my throat.

183
00:14:58.380 --> 00:15:00.000
Of course, that's not true.

184
00:15:00.460 --> 00:15:23.400
That's a false prediction, but I'll hold it in my mind until I go, I pour myself a glass of water and take a drink, then I can release the prediction. Or if I turn the burner on on the stove, I'll hold the prediction that I need to turn off the burner until I turn off the burner. So I sort of keep track of the state of things by active inference.

185
00:15:23.400 --> 00:15:24.800
InScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScriptScript

186
00:15:24.800 --> 00:15:28.940
So yeah, this is a really cool, cool framework.

187
00:15:29.140 --> 00:15:31.780
Sorry, I was going to say,

188
00:15:31.820 --> 00:15:33.360
so the Burnham one was super interesting.

189
00:15:33.520 --> 00:15:36.520
So there you said, once you've turned it on,

190
00:15:37.140 --> 00:15:39.960
you would have like in your present mind,

191
00:15:40.060 --> 00:15:42.600
like a prediction that you need to turn it off

192
00:15:42.600 --> 00:15:45.060
rather than predicting that it will turn off.

193
00:15:45.200 --> 00:15:48.540
Yeah, there's some kind of accounting system in the brain

194
00:15:49.680 --> 00:15:53.640
where, okay, you know, I know now I'm turning this on

195
00:15:54.900 --> 00:15:58.960
to sort of hold this state in my brain.

196
00:15:59.340 --> 00:16:02.780
And I know that I can stop holding the state

197
00:16:02.780 --> 00:16:03.920
once I turn it off.

198
00:16:04.300 --> 00:16:07.400
So we can sort of frame that as what I'm holding

199
00:16:07.400 --> 00:16:10.060
is the prediction that I, you know, the burner is off.

200
00:16:10.200 --> 00:16:10.660
Gotcha.

201
00:16:10.800 --> 00:16:11.820
Right, yeah.

202
00:16:12.300 --> 00:16:13.760
I have a question about this

203
00:16:13.760 --> 00:16:17.300
because you say that making a prediction,

204
00:16:17.920 --> 00:16:20.660
for example, that you are not thirsty

205
00:16:20.660 --> 00:16:22.800
is said to be a false prediction

206
00:16:22.800 --> 00:16:23.720
according to this framework.

207
00:16:24.220 --> 00:16:24.780
Right.

208
00:16:24.800 --> 00:16:27.760
Whereas I typically interpret my desire

209
00:16:27.760 --> 00:16:31.740
to grab a glass of water to be a type of wanting.

210
00:16:31.860 --> 00:16:32.800
It's a type of desire

211
00:16:32.800 --> 00:16:35.960
rather than thinking of it as a false prediction,

212
00:16:36.120 --> 00:16:37.940
which I hold true until it becomes true

213
00:16:37.940 --> 00:16:39.480
by way of my acting in the world.

214
00:16:39.660 --> 00:16:41.320
And is there a difference?

215
00:16:41.340 --> 00:16:43.640
Is there a difference between like falsely predicting

216
00:16:43.640 --> 00:16:47.200
and simply wanting or desiring?

217
00:16:47.240 --> 00:16:49.520
Or are these like functionally the same thing?

218
00:16:49.640 --> 00:16:52.540
So it's the same phenomena, same mechanism for achieving.

219
00:16:52.860 --> 00:16:54.300
Yeah, that's a good question.

220
00:16:54.800 --> 00:16:56.680
I would say that with it,

221
00:16:56.720 --> 00:16:58.920
it of course depends on the mechanism.

222
00:16:59.300 --> 00:17:01.600
And with some mechanisms,

223
00:17:02.140 --> 00:17:04.820
it's implemented in such a way that they would be the same.

224
00:17:04.900 --> 00:17:07.140
And in some mechanisms, it would not be.

225
00:17:07.200 --> 00:17:09.720
So I guess this would be an argument

226
00:17:09.720 --> 00:17:11.600
for sort of dropping down

227
00:17:11.600 --> 00:17:13.820
and paying attention to implementation,

228
00:17:14.220 --> 00:17:16.280
which is the third domain, of course.

229
00:17:16.440 --> 00:17:17.200
Right, cool.

230
00:17:17.920 --> 00:17:23.020
So can you tell us more about Buddhist conceptions of Tana

231
00:17:23.020 --> 00:17:24.780
and how that works?

232
00:17:24.800 --> 00:17:27.520
And how that might have influenced your development

233
00:17:27.520 --> 00:17:31.240
of your framework in Veso computational theories?

234
00:17:32.000 --> 00:17:32.820
Yeah, sure.

235
00:17:34.100 --> 00:17:38.480
So I want to give a shout out to both Romeo Stevens

236
00:17:38.840 --> 00:17:40.240
and Nick Camerata.

237
00:17:41.180 --> 00:17:44.160
They've really done a lot of foundational work

238
00:17:44.160 --> 00:17:49.000
on thinking about Tana, novel translations of Tana,

239
00:17:49.120 --> 00:17:51.420
or like sorting through, okay, what's a good translation

240
00:17:51.420 --> 00:17:52.380
and what's not.

241
00:17:52.520 --> 00:17:54.460
And in first person,

242
00:17:54.800 --> 00:17:59.240
exploration that, you know, can we see this?

243
00:17:59.380 --> 00:18:03.680
Can we put a timeframe on, you know,

244
00:18:03.680 --> 00:18:07.060
does it happen within 25 to 50 milliseconds

245
00:18:07.060 --> 00:18:08.840
or a hundred milliseconds or whatnot?

246
00:18:09.520 --> 00:18:12.800
So they've done really, really great work.

247
00:18:14.640 --> 00:18:19.840
I think, so my focus is generally

248
00:18:19.840 --> 00:18:22.860
sort of connecting these domains and saying,

249
00:18:22.940 --> 00:18:24.520
okay, this Buddhist,

250
00:18:24.800 --> 00:18:30.740
Tana is sort of using this active inference system

251
00:18:31.280 --> 00:18:32.680
in an unskillful way,

252
00:18:33.760 --> 00:18:39.100
trying to control sensations in ways that don't make sense

253
00:18:39.100 --> 00:18:43.200
in some sort of like there are in programming,

254
00:18:43.340 --> 00:18:46.260
we can talk about type errors, trying to, you know,

255
00:18:47.320 --> 00:18:51.300
add two letters together or multiply two letters together

256
00:18:51.300 --> 00:18:52.080
or something like that.

257
00:18:52.140 --> 00:18:54.780
It's just like, that's not the sort of thing that we're,

258
00:18:56.800 --> 00:19:03.420
the tanha as unskillful active inference frames as the brain engages in a lot of that stuff

259
00:19:03.420 --> 00:19:10.840
when it tries to manipulate sensations that it's it's often unskillful in the ways that it tries to

260
00:19:10.840 --> 00:19:18.960
make predictions about sensations and that in a in a moment by moment sense this happens a lot

261
00:19:18.960 --> 00:19:27.640
and this really adds up to like uh a lot of un sort of submerged unpleasantness in your moment

262
00:19:27.640 --> 00:19:33.480
to moment experience and that i mean the the buddhists say that and if you fix this if the

263
00:19:33.480 --> 00:19:39.940
the vipassana frame is you can see this happening and once you see it you can't really unsee it

264
00:19:39.940 --> 00:19:48.300
and you'll start to not do wrong things right so when you say see it what do you mean by that do

265
00:19:48.300 --> 00:19:48.780
you mean like a

266
00:19:48.960 --> 00:19:57.140
literal seeing um i would say that uh when you observe what the mind is doing i wouldn't say uh

267
00:19:57.140 --> 00:19:59.880
it necessarily flashes into your visual

268
00:20:00.000 --> 00:20:06.420
field, but I do think that everyone's going to be a little bit different in how it presents.

269
00:20:07.260 --> 00:20:17.260
But I would say that I think the takeaway would be sort of a bit of a shock, like, oh, wow, I'm doing that all the time?

270
00:20:17.260 --> 00:20:39.140
This feels suboptimal. And I think that progress on the path, on sort of getting better at not doing this, is sort of developing a taste for what it feels like, and a distaste for like, okay, no thanks. Like, I don't need to do that to myself.

271
00:20:39.620 --> 00:20:44.440
In your blog post, you use the word stress, just in a passing part of it.

272
00:20:44.520 --> 00:20:47.240
Is that potentially one of the ways that you're going to be able to do that?

273
00:20:47.260 --> 00:20:52.260
Yeah, I think that maybe someone would experience this. So if you're grasping a lot, partly that might add to your stress.

274
00:20:52.520 --> 00:21:01.480
Yeah, so I sort of break down the Tanha as unskillful active inference into three buckets.

275
00:21:02.460 --> 00:21:12.400
The first bucket is just if you try to control your sensations or control your environment more than you have the energy to.

276
00:21:12.540 --> 00:21:17.240
And, you know, we all know people that try to control their sensations.

277
00:21:17.260 --> 00:21:24.440
And we can all kind of feel, okay, you know, that runs into problems.

278
00:21:25.240 --> 00:21:29.060
And you just get stressed trying to control everything, and it doesn't work.

279
00:21:30.000 --> 00:21:34.960
I think what the Buddhists are pointing to, though, is a little bit different.

280
00:21:35.780 --> 00:21:46.560
It's that we try to apply active inference or predictions to sensations in ways that don't make sense, could never come true.

281
00:21:47.260 --> 00:21:51.280
Or just are sort of bound to cause us a lot of suffering.

282
00:21:51.580 --> 00:21:51.920
A couple of examples?

283
00:21:52.240 --> 00:22:02.040
Yeah, so a couple simple examples would be like if a sensation is good, it's a nice, pleasant, tasty sensation.

284
00:22:02.680 --> 00:22:11.820
For example, if you're eating some cake and you really like it and you make the prediction, I will have this taste in my mouth forever.

285
00:22:12.740 --> 00:22:17.240
Of course, you're going to run into trouble because, okay, you're going to have this problem.

286
00:22:17.260 --> 00:22:22.040
And then you're going to eat all the cake, trying to eat that taste in your mouth, and then you're going to be out of cake.

287
00:22:23.180 --> 00:22:25.300
And then you're just going to be left with tension.

288
00:22:25.640 --> 00:22:27.800
You can't make that prediction true.

289
00:22:28.080 --> 00:22:32.660
And you don't have sort of any easy way to release that tension.

290
00:22:33.220 --> 00:22:39.640
The word forever in there is very powerful because, yeah, I wouldn't have thought I feel like I want the taste forever.

291
00:22:39.840 --> 00:22:43.440
But maybe there is an aspect to that of the desire of the taste.

292
00:22:43.860 --> 00:22:44.260
Yeah.

293
00:22:44.360 --> 00:22:47.080
So Romeo Stevens says this.

294
00:22:47.260 --> 00:22:54.420
This nice frame that we try to make our sensations stable, controllable, and satisfying.

295
00:22:55.050 --> 00:23:03.920
And, of course, the Buddhists would say, you know, we can never ensure any sensation stable or controllable or satisfying.

296
00:23:05.000 --> 00:23:09.320
So we're sort of doomed to discomfort when we assume that.

297
00:23:10.070 --> 00:23:10.400
Right.

298
00:23:10.560 --> 00:23:12.360
Going back, do you...

299
00:23:12.360 --> 00:23:16.640
So you mentioned before that there are three unskillful active influences.

300
00:23:16.640 --> 00:23:17.240
Yeah.

301
00:23:17.260 --> 00:23:22.320
And the first one was that if you try to control the environment more than you have energy to, that leads to stress.

302
00:23:22.500 --> 00:23:26.340
What are the other two causes of stress or unskillful?

303
00:23:26.480 --> 00:23:26.860
Yeah.

304
00:23:27.260 --> 00:23:30.120
So the first is just trying to control too much.

305
00:23:30.460 --> 00:23:35.040
The second was trying to control in ways that don't make sense.

306
00:23:35.180 --> 00:23:38.600
For example, you know, I'll have this taste forever.

307
00:23:39.560 --> 00:23:45.540
Or, you know, if you drop a drink on the floor and you make the prediction, that didn't happen.

308
00:23:45.700 --> 00:23:46.180
Right.

309
00:23:47.680 --> 00:23:49.980
So it's not going to make it true.

310
00:23:50.080 --> 00:23:56.120
You can hold it, but it's not the appropriate prediction for the event.

311
00:23:56.620 --> 00:24:05.680
And then the third bucket or category of sort of ways that we can be unskillful in using active inference is...

312
00:24:05.680 --> 00:24:08.680
And I think this is really a cool, important thing.

313
00:24:08.800 --> 00:24:11.080
It's context desynchronization.

314
00:24:11.360 --> 00:24:11.800
Mm-hmm.

315
00:24:11.800 --> 00:24:14.720
So if you're doing a very hard math problem.

316
00:24:14.720 --> 00:24:28.320
And you're sort of holding, you know, using tension as medium-term memory and sort of holding a problem in your mind by sort of let's pinch here and let's hold here and so on.

317
00:24:28.400 --> 00:24:38.100
And then someone knocks on the door and asks to, you know, get a cup of flour, a cup of sugar, whatever is the Xanadu scenario.

318
00:24:39.500 --> 00:24:43.940
Or, you know, someone says, oh, you know, we have an emergency.

319
00:24:44.720 --> 00:24:45.740
There's a fire alarm or something.

320
00:24:45.820 --> 00:24:49.020
The context gets switched very rapidly.

321
00:24:49.280 --> 00:24:49.880
Mm-hmm.

322
00:24:49.880 --> 00:25:00.000
The tension that you're holding in your mind to remember the things in context one no longer has sense in context two.

323
00:25:00.140 --> 00:25:20.700
it's just stress. It's just tension. It's just suffering. It only has meaning within the right context. And it's like the, you know, if you're working on a computer program and the memory allocation changes, then your pointers are not going to point to the right things.

324
00:25:20.700 --> 00:25:34.700
And I think this actually happens a lot. And that a lot of our stress, it's sort of these fragments of tension that made sense in one context, but now we're in a new context where they don't make sense.

325
00:25:35.180 --> 00:25:43.400
Okay. So what is the, that's what unskillful, active, and looks like. What does skillful management tension look like?

326
00:25:43.660 --> 00:25:46.420
Right. Not using it a lot.

327
00:25:47.720 --> 00:25:48.860
Which means letting go?

328
00:25:50.700 --> 00:26:08.840
Just kind of, you know, a little bit, a touch here, a touch there. We still need to navigate the world. And we may need tension to do that. We may need predictions to do that. We may need active inference to do that. But don't, like, make as few as possible. And don't hold them.

329
00:26:09.360 --> 00:26:14.220
So skillfully choose your predictions and then like update them more often or replace them?

330
00:26:14.440 --> 00:26:17.700
Yeah, yeah. Just hold them gently.

331
00:26:17.920 --> 00:26:18.360
Right.

332
00:26:18.420 --> 00:26:20.680
So how does this now like lead to the next step?

333
00:26:20.680 --> 00:26:22.860
to predictions?

334
00:26:23.920 --> 00:26:23.960
Yeah.

335
00:26:24.220 --> 00:26:32.400
And what is like the optimal level prediction or active inference to do in order to have a healthy relationship, internal states, whether it be?

336
00:26:32.500 --> 00:26:34.840
Right. That's a good question.

337
00:26:35.840 --> 00:26:50.100
I think that, you know, as we were saying, you know, being sparing with the predictions that understand that a prediction is a commitment to tension and a little bit of suffering until it comes true.

338
00:26:51.160 --> 00:26:53.520
But you also need them to navigate the world.

339
00:26:53.920 --> 00:26:54.080
Yeah.

340
00:26:54.080 --> 00:27:09.300
I think that, so we were all at just camp together where we met. And Vin and I were talking about the self and, you know, the self as kind of the sum total of tension in the mind.

341
00:27:09.380 --> 00:27:23.900
And I think this sort of nicely dovetails with some of the ways the Buddhists talk about the self and tension as sort of not the optimal way to navigate the world.

342
00:27:25.580 --> 00:27:29.820
Um, so I guess I would just say that, uh, you know, everything is contextual.

343
00:27:30.060 --> 00:27:33.300
There may not be a, you know, this is always good.

344
00:27:33.480 --> 00:27:38.200
This is always bad, but there will always be a dimension of, is this skillful?

345
00:27:38.260 --> 00:27:39.140
And is this not?

346
00:27:39.200 --> 00:27:42.160
Can you talk about what, what good and bad mean in this concept?

347
00:27:42.300 --> 00:27:44.020
I know that's like, right, right.

348
00:27:44.160 --> 00:27:47.040
Um, yeah, that's, that's a hard problem.

349
00:27:47.400 --> 00:27:54.620
Um, I mean, so in my more formal philosophy research, um, I study pain and pleasure.

350
00:27:55.360 --> 00:28:00.340
And I'd be, I'd be hesitant to, to say, yes, you know, pleasure is always good.

351
00:28:00.420 --> 00:28:01.200
Pain is always bad.

352
00:28:01.360 --> 00:28:06.220
Uh, because now sometimes these sensations are trying to do things for us.

353
00:28:06.360 --> 00:28:12.240
And, um, I think that, uh, I mean, good and bad could be framed in terms of the free energy

354
00:28:12.240 --> 00:28:12.640
principle.

355
00:28:13.060 --> 00:28:20.480
Um, and sort of, it could, you know, be framed in terms of, uh, adaptiveness to the environment

356
00:28:20.480 --> 00:28:21.420
and so on.

357
00:28:21.460 --> 00:28:25.220
So I'm not sure I have a, a sort of clean.

358
00:28:25.360 --> 00:28:27.180
Like this is what good is.

359
00:28:27.400 --> 00:28:29.500
This is what bad is, but just an appreciation.

360
00:28:30.140 --> 00:28:31.300
There's, there's a lot of theories.

361
00:28:31.400 --> 00:28:31.680
Yeah.

362
00:28:31.760 --> 00:28:32.820
This is a lot of context.

363
00:28:32.880 --> 00:28:33.160
Yeah.

364
00:28:33.280 --> 00:28:35.260
That's a pretty tall order to deliver.

365
00:28:35.580 --> 00:28:40.500
Um, anyway, uh, just moving on from that, I suppose, um, I really want to talk about

366
00:28:40.500 --> 00:28:46.720
qualia and when you worked in a institute that had qualia in its name, and I was wondering

367
00:28:46.720 --> 00:28:50.980
what you thought, um, what, what do you, what do you take qualia to?

368
00:28:51.020 --> 00:28:52.180
And can you tell us more about that?

369
00:28:52.300 --> 00:28:52.640
Yeah.

370
00:28:53.180 --> 00:28:55.240
Um, so I always.

371
00:28:55.360 --> 00:28:59.440
Uh, introduce quality as the components of subjective experience.

372
00:29:00.320 --> 00:29:06.140
And it's a little bit, you know, uh, the, the easy, uh, explanation would be, you know,

373
00:29:06.140 --> 00:29:12.040
there's red, there's heavy, there's, uh, hot, et cetera.

374
00:29:12.420 --> 00:29:18.360
Um, but what the actual natural kinds of qualia are is a very interesting unsolved question.

375
00:29:18.640 --> 00:29:19.640
Um, thanks Mike.

376
00:29:19.780 --> 00:29:20.220
Yeah.

377
00:29:20.320 --> 00:29:20.700
You bet.

378
00:29:20.780 --> 00:29:21.460
Uh, thanks Ben.

379
00:29:21.580 --> 00:29:22.740
And, uh, thanks Francis.

380
00:29:22.880 --> 00:29:23.540
Thank you.

381
00:29:23.560 --> 00:29:24.340
It was super interesting.

382
00:29:25.360 --> 00:29:25.580
Yeah.

383
00:29:25.700 --> 00:29:30.080
They're connecting to some really important things about how we experience our life every

384
00:29:30.080 --> 00:29:30.320
day.

385
00:29:30.420 --> 00:29:33.700
Um, but also a theory that connects them up, which is great.

386
00:29:34.900 --> 00:29:35.420
Yeah.

387
00:29:36.440 --> 00:29:36.960
Yeah.