Those Conversations

When evaluating a conversation, many criteria can describe it. Some are serious, funny, sad, or happy. What shapes a conversation’s feel? Is it the semantic content or the environment where it takes place? Both probably contribute, but I would argue that there is something more— something that cannot (at least with current technology) be measured: the level of connection between the two participants. But what is an “amount of connection”? Again, I think that many components go into this calculation, but I will loosely define two key elements. i) Semantic Alignment: Shared perspective and understanding on a situation. ii) Bandwidth: The amount of information that can be transferred from one mind to another.

It is essential to note that these properties are interdependent. To help conceptualize how these properties affect a conversation, here are two examples: i) You have a disagreement with a family member, it starts with something small, but after a few snappy remarks, escalates into a small yelling match, and you both leave the conversation feeling hurt, unheard, and in sum, ‘bad’. What made the conversation ‘bad’? You had one perspective on the situation, your family member had another, and throughout the conversation, this discrepancy grew as emotions clouded your reason.

In other words, a semantic disalignment. Closely related, the bandwidth of the conversation was extremely low; the information you were trying to convey to your family member wasn’t getting through. ii) You have a disagreement with a family member, but at the beginning of the conversation, you work to establish a ground truth about the disagreement. Through this initial work, you both realize that you did not fully understand the situation. After reconciling the differences, you walk away from the interaction feeling good, and you have an understanding with the other person.

In this situation, there was a consistent interpretation of the semantic content (you both came to an understanding), and there was high bandwidth (you both were able to communicate your experience fully), with the information being sent from one party to another being received as a result of active listening and empathy. If misalignment drags us into conflict, and alignment restores harmony, what happens when alignment and bandwidth both approach their peak? I will apply this framework to describe a specific type of conversation. This kind of conversation does not have a name, but those who have had similar discussions will understand from my description.

I will refer to this type of conversation as those conversations throughout the rest of this essay. Those conversations are the type that you walk away from feeling energized, inspired, and ‘good’. You felt like you truly connected to and understood the other party, and you felt like they reciprocated. It felt natural, the ideas kept building, and your trains of thought aligned into a beautiful symphony of shared experience. In other words, the interpretation of semantic content was almost perfectly consistent, and the bandwidth of the conversation was so high that at times, words were unnecessary for the achievement of consistency.

A vague reference (i.e., those conversations) is sufficient for the complete transfer of information and understanding between parties (i.e., the experience of remembering and relating to those conversations). Those conversations are especially interesting, as what I described above is nearly telepathy. The bandwidth of communication is exceptionally high with minimal (if not negligible) utilization of communication modalities. What is more interesting, though, is the neural activity that underlies this achievement, and how the computation enables synchronization across a shared experience.

And if there is a way to consciously induce such communication through high-bandwidth braincomputer interface (BCI) encoding and decoding technology, thereby achieving real telepathy. Many limitations need to be overcome for BCIs to achieve functional telepathy (the majority of which involve physics-based interfacing problems and ethical dilemmas); this writing will disregard the physical boundaries and instead focus on the informational aspects of telepathy, and how the encoding of incoming information will need to transcend semantic-based methods to encode conscious experience itself.

To the best of our knowledge, information in the brain is represented through specific patterns of neural firing, primarily dictated by external stimuli and the weights of learned synaptic connections. By recording particular properties of neural computation, we can decode specific information. The goal of a telepathy-based BCI is to record a distinct pattern of activity in one brain, transmit it to another brain, and reencode it for the conscious experience of the information. However, for this medium of communication to be adopted, it must be superior to our current forms of communication.

In other words, the bandwidth of communication must surpass the bandwidth of communicating the same expression via words, facial expressions, etc. What does that higher bandwidth look like? When speech works, a speaker compresses the experience into a single statement, sending it via changes in air pressure, and then the receiver uncompresses it into the experience itself. Telepathy would demand decoding a given experience from a sender’s neural patterns, compressing that code for efficient transfer, and encoding it into the corresponding (but vastly different) representation in the receiver’s brain.

The most obvious answer is to map experiences to how they are represented as words in the receiver’s brain, but then, outside of medical applications, why not just use your spoken word instead? To beat speech in bandwidth, a telepathic BCI must transmit the experience itself—the neural firing pattern that gives rise to the conscious feeling—and recreate that pattern in the receiver’s brain, not just send the neural code for the words that label the experience. Many non-interfacing hurdles remain, for example: how can we replicate the same experience between two people with vastly different neural wiring?

How can we minimize latency and maximize bandwidth? What would a user interface even be for this? And many ethical considerations: How can we protect the data being sent and received? How do we discriminate between intentional communication and internal monologue? How will this affect the human psyche? However, if I delve into those issues, this would be far too lengthy, and no one would read it—and quite frankly, neither I nor the scientific community knows the answer to these. To wrap up, I’ll try to describe what I imagine the experience of communicating using this medium to be like and forecast some societal implications.

The experience of communicating in this way is challenging to conceptualize, but I think the best way to do this is to consider, again, those conversations. I imagine that this novel form of communication would feel like one of those conversations, except, not even a small phrase—or word—is needed for the understanding to take place between the two parties. It would feel like your mind is genuinely connected to the receiver’s, and vice versa. Instead of you thinking of or hearing the words those conversations, and then experiencing the semantic content as a result, you would experience the semantic content directly.

I believe that when using this medium to communicate, disagreements can be addressed with a whole-perspective interpretation of the situation, maximum bandwidth, and empathy-driven solutions, as seen in the second example conversation I provided. Resulting from this? A harmonious society, united by relation, that solves its disputes through empathy, shares its innovations through experience, and spreads ideas through experience alone. I would be a fool to think I will be able to experience this in my lifetime, but I would be a coward if I didn’t give my all to try.

Here’s to the future, my friends. Let’s try our best not to destroy ourselves before we get there.