“User: My boyfriend made me come here.
Chatbot: Your boyfriend made you come here.
User: He says I’m depressed much of the time.
Chatbot: I am sorry to hear that you are depressed.
User: It’s true. I am unhappy.
Chatbot: Do you think coming here will help you not to be unhappy?”
This is a fragment of a conversation between a patient and a therapist chatbot. In fact, we are talking about the first chatbot ever created. Its name is ELIZA and it was developed in 1966, by Joseph Weizenbaum, a computer scientist at MIT (Massachusetts Institute of Technology). Even if this computer program was made as a satire initially, it was then found both useful and captivating by many people. Even doctors saw it as a potentially transformative tool. They believed that several hundred clients - in need of mental healthcare - an hour, could be handled by a computer system designed for this purpose. It was the beginning of a new era.
Nowadays, technology is making advancements at a rapid pace, with ChatGPT becoming a global phenomenon and gaining substantial popularity in the past year. But where did it all start?
As I stated earlier, the first chatbot in the history of Computer Science is ELIZA, designed to mimic human conversation and further used to simulate psychotherapy. She would interact with users in typed conversations and would respond to their affirmations or questions based on pre-made scripts that were in the program. If the bot did not quite understand what the person said, or how to reply to it, it would fall back on generic phrases like “that’s very interesting” and “go on.”. Even if Weizenbaum, the scientist from MIT and ELIZA’s creator, didn’t initially intend for it to be utilised as a form of mental healthcare, he still went and developed a script meant to simulate psychotherapy. Therefore, the chatbot is recognised for the innovative way of substituting a human being.
What’s so gripping about it is that ELIZA didn’t receive any specialised training or programming in psychotherapy, yet it could still be effective when it comes to providing aid to people in need. Its generic text outputs were giving the illusion of understanding people’s dilemmas and intentions, by mirroring users’ language back at them. This is one smart technique used in therapy, as the mere repetition of a fact said by someone from the outside can lead you to significant realisations about yourself and about your mind. We might notice when something is triggering us, but we don’t always ask ourselves what’s the reason for the way we feel. Therefore, a radical question coming from someone else, such as: “What has led you to this disposition?”, “Would you rather react differently to this event if you could go back in time? ”, could be helpful, especially if one does not practise introspection too often. By introspection we refer to the analysis of one’s conscious thoughts and feelings, the habit of observing the current mental state.
Now, it’s true that by today's standards, ELIZA fails very quickly if you ask it a few complex questions. She would repeat the same phrases over and over again or inquire more about its user, by saying: “Tell me more…”, “Are you sure about that?”. Students and researchers are nowadays delighted to make their own versions of this program by revising and adding to the code and scripts, ameliorating the performances of this chatbot.
Some people believe that such programs could never fully substitute human beings in the field of mental healthcare, where emotional intelligence, empathy and human connection are fundamental. Nevertheless, we ought to appreciate this ingenious technique, which could be tried by all of us from the comfort of our homes.
Bibliography:
Khullar D. (2023, February 27). Can A.I. Treat Mental Illness? The NewYorker.
Law M. (2022, december 07). From ELIZA to ChatGPT: The evolution of chatbots technology. Technology Magazine
Glover E. (2023, July 14). What Is the Eliza Effect? Built In
Try a simulation of ELIZA by using this link: https://web.njit.edu/~ronkowit/eliza.html
Comments