I like the idea! If you have it, the question is when does it start? Let’s look at it for different aspects of consciousness:
Sentience (Bentham, Singer): Behavioral responses to pleasure or pain stimuli and physiological measures. This is observable across animal species, from mammals to some invertebrates and it should be known when responses to such stimuli start in the embryo.
Wakefulness: Measureable in virtually all animals with a central nervous system by physiological indicators such as EEG, REM, and muscle tone. The fetus is known to have a sleep wake rhythm, but I don’t know when it starts.
Dennet’s Intentionality: Treating living beings as if they have beliefs and desires makes good predictions for many animal species, esp. social, like primates, cetaceans, and birds. Infants show goal directed behaviors right after birth. I remember ultrasound photos that show babies suckling their thumb. I think we can identify when the nervous system is first capable of goal direction.
Dehaene’s Phenomenal Consciousness: A perception or thought is conscious if you can report on it. As this requires language or measuring neural patterns that are similar to humans during comparable reports, I think this starts when communicable representstions of perceptions first form, for toddlers around age one at the earliest with baby sign language.
Gallup’s Self-Consciousness: Recognition of oneself e.g. in a mirror. Requires sufficient sensual resolution and intelligence for a self-model. Dito.
Rosenthal’s Meta-Consciousness: This is investigated through introspective reports on self-awareness of cognitive processes or self-reflective behaviors. Requires more abstraction. Maybe at age five?
I’d like to explore these in more depth, but for now I’ll just reduce all the angles you provided to the helpful summaries/applications you provided. I’ll call the perspective of going from adult human to zygote the “physical history” and the perspective of going up the ancestral tree as the “information history” (for simplicity, maybe we stop as soon as we hit a single-celled organism).
Sentience: This feels like a continuous thing that gets less and less sophisticated as we go up the information history. In each generation, the code gets a little better at using the laws of physics and chemistry to preserve itself. Of course if one has a threshold for what counts as sentience, it will cross it at some point, but this still strikes me as continuous.
Wakefulness: This would strike me as a quantized thing from both the information and physical history perspective. At some point in both histories, the organism/cell would pick up some cyclic behavior.
Intentionality: I’d need to look more at this, because my interpretation of your first sentence doesn’t make sense with the second.
Phenomonal, Self-Consciousness, Meta-Consciousness: Definitely quantized in both perspectives
When I was thinking of subjective experience, I think the only concepts here that are either weaker or stronger than what I had in mind are the last two. For the rest, I think I can both imagine a robot that satisfies the conditions and imagine a conscious being that does not satisfy the condition.
But the last two still feel too strong. I will think more about it.
Sentience: This feels like a continuous thing that gets less and less sophisticated as we go up the information history. In each generation, the code gets a little better at using the laws of physics and chemistry to preserve itself.
I think that getting better at using the laws of physics to reproduce, is some stage before sentience. Sentience as defined by Singer is about responses to pleasure and pain stimuli—which is a specific adaptation that requires specific neural pathways that are not present, e.g., in bacteria. I’m fine with adding another layer before sentience, let’s call it reproduction, and maybe that one is continuous as you suggest, but it stretches what people call conscious. Sure, you can define consciousness to include that layer, and maybe that is what people call panpsychism, but to me, that seems more like expanding a definition by applying an affect heuristic.
But the last two still feel too strong. I will think more about it.
For what it’s worth, I think we will soon see “robots” or LLMs or some such systems that have meta-consciousness or self-consciousness. There are reports of LLMs passing the mirror test and if they can do that and argue the case—and I have seen pretty advanced arguments about reflection too—then you have meta-consciousness also.
I like the idea! If you have it, the question is when does it start? Let’s look at it for different aspects of consciousness:
Sentience (Bentham, Singer): Behavioral responses to pleasure or pain stimuli and physiological measures. This is observable across animal species, from mammals to some invertebrates and it should be known when responses to such stimuli start in the embryo.
Wakefulness: Measureable in virtually all animals with a central nervous system by physiological indicators such as EEG, REM, and muscle tone. The fetus is known to have a sleep wake rhythm, but I don’t know when it starts.
Dennet’s Intentionality: Treating living beings as if they have beliefs and desires makes good predictions for many animal species, esp. social, like primates, cetaceans, and birds. Infants show goal directed behaviors right after birth. I remember ultrasound photos that show babies suckling their thumb. I think we can identify when the nervous system is first capable of goal direction.
Dehaene’s Phenomenal Consciousness: A perception or thought is conscious if you can report on it. As this requires language or measuring neural patterns that are similar to humans during comparable reports, I think this starts when communicable representstions of perceptions first form, for toddlers around age one at the earliest with baby sign language.
Gallup’s Self-Consciousness: Recognition of oneself e.g. in a mirror. Requires sufficient sensual resolution and intelligence for a self-model. Dito.
Rosenthal’s Meta-Consciousness: This is investigated through introspective reports on self-awareness of cognitive processes or self-reflective behaviors. Requires more abstraction. Maybe at age five?
I’d like to explore these in more depth, but for now I’ll just reduce all the angles you provided to the helpful summaries/applications you provided. I’ll call the perspective of going from adult human to zygote the “physical history” and the perspective of going up the ancestral tree as the “information history” (for simplicity, maybe we stop as soon as we hit a single-celled organism).
Sentience: This feels like a continuous thing that gets less and less sophisticated as we go up the information history. In each generation, the code gets a little better at using the laws of physics and chemistry to preserve itself. Of course if one has a threshold for what counts as sentience, it will cross it at some point, but this still strikes me as continuous.
Wakefulness: This would strike me as a quantized thing from both the information and physical history perspective. At some point in both histories, the organism/cell would pick up some cyclic behavior.
Intentionality: I’d need to look more at this, because my interpretation of your first sentence doesn’t make sense with the second.
Phenomonal, Self-Consciousness, Meta-Consciousness: Definitely quantized in both perspectives
When I was thinking of subjective experience, I think the only concepts here that are either weaker or stronger than what I had in mind are the last two. For the rest, I think I can both imagine a robot that satisfies the conditions and imagine a conscious being that does not satisfy the condition.
But the last two still feel too strong. I will think more about it.
I think that getting better at using the laws of physics to reproduce, is some stage before sentience. Sentience as defined by Singer is about responses to pleasure and pain stimuli—which is a specific adaptation that requires specific neural pathways that are not present, e.g., in bacteria. I’m fine with adding another layer before sentience, let’s call it reproduction, and maybe that one is continuous as you suggest, but it stretches what people call conscious. Sure, you can define consciousness to include that layer, and maybe that is what people call panpsychism, but to me, that seems more like expanding a definition by applying an affect heuristic.
I’m not sure what “the last two”. :confused:
*The last two bullet points. Meta-consciousness and self-consciousness
For what it’s worth, I think we will soon see “robots” or LLMs or some such systems that have meta-consciousness or self-consciousness. There are reports of LLMs passing the mirror test and if they can do that and argue the case—and I have seen pretty advanced arguments about reflection too—then you have meta-consciousness also.