Full Text Article

Introduction

Emotion - what does this word mean? According to the Merriam-Webster dictionary, the word emotion refers to “…a conscious mental reaction (such as anger or fear) subjectively experienced as strong feeling usually directed toward a specific object and typically accompanied by physiological and behavioral changes in the body…” [1].

AI and emotion have forever been a taboo when spoken about together. AI is often pitted as unable to experience subjective emotion, and therefore unable to provide inherently personalized and empathetic responses. Rather, AI has been acknowledged as a machine device used to produce efficiency and optimize production. As [2] infer in their book “How Reason Almost Lost Its Mind”, the purpose of computer machines is to rationalize the unthinkable, to make ruthless decisions that humans are rather unable to base on felt emotions (p.16). However, when contemplating the Netflix original series “Too Hot To Handle”, this idea that AI mechanisms hold the ineptitude to experience and deliver personalized and sympathetic responses can be re-imaged to indeed actually possibly retain the intelligence and capacity to perhaps do so. For when analyzing this Netflix show, it is observable that AI robots can perhaps act as this sort of robotic therapist, offering possibly higher understanding and agility in effecting change towards problematic emotional behaviors as showcased by the various contestants.

When grappling with the concept of emotions, AI technologies and therapeutic attempts, the notion of a “cold hard-drive” installed within a computer operated system versus a warm human heart housed within a human being, rises multiple questions. Queries such as “does the use of AI technologies provide the opportunity to avoid human contact and alleviation possible sentiments of being judged or embarrassed by a human therapist decrease?”, “Do AI technologies bring about the ability for people to talk more freely, feel less restricted, offer greater anonymity, and provide more emotional security?”, “Do the responses from AI technology provide a better sense of unbiased therapy?”. With that said, this paper seeks to investigate if AI technology led therapy offers a preferred medium of therapy versus human-led therapy within contemporary society today.

Literature Review

What is AI? “Artificial intelligence is a constellation of many different technologies working together to enable machines to sense, comprehend, act, and learn with human-like levels of intelligence…” [3]. AI is also described to exists in varying ‘degrees’ - Narrow (or “weak”) AI and General (or “strong”) AI. Narrow AI is defined as artificial intelligence that accompanies us in our everyday lives, performing single tasks and distributing efficiency in our day-to-day (examples of this type of AI include weather applications, digital assistants, etc.,) [3]. Narrow AI in sum is constructed and utilized for specific or limited tasks, it is a type of analytical intelligence that defines functions and applications [4]. General (or “strong”) AI refers to artificial intelligence that relates to the common perception of what AI constitutes (i.e., machines that provide human-like intelligence which thinks strategically and creatively, all whilst handling more complex tasks) [3]. General AI therefore refers to the type of artificial intelligence that is capable of finding a solution without human intervention [4]. According to [3], “…artificial intelligence remains an extension of human capabilities, not a replacement.”.

AI Intelligence/Rationality

The term rational by definition refers to “Having reason or understanding.” (Merriam-Webster Dictionary, online). The ability to decide rationally is possessed in the strength and competence to think responsibly [5]. According to [5], AI devices represent intelligent agents, which refers to:

“An intelligent agent “takes the best possible action in a situation” (p. 30), i.e. it is a rational agent the one which, for each possible percept sequence, is supposed to “select an action that is expected to maximize its performance measure, given the evidence provided by the percept sequence and whatever built-in knowledge the agent has” (p. 37).

AI therefore is able to engage in rational decision-making processes based on algorithms programmed by the machine’s primary intelligence source (i.e., the human). However, the AI devices primary source of intelligence rests as potentially being biased, for all human intelligence is sourced from a biased perspective. That said, the computer machine resorts to producing a “rational decision” based on algorithmic data sets leading to said decision [5]. Be that as it may, is AI therefore then inherently and purely rational in its nature?

[6] further elaborate this dilemma by stating that AI refers to a system “…designed by humans, interpreting the already collected data and decision the best action to take, according to the pre-defined parameters, in order to achieve the given goal” (p.58455). The authors note an important aspect whereby the notion of rationality founded in AI’s decision-making process, which refers to the actions of interpreting and deciding on the best action to take given the pre-determined algorithms, thus illustrating the rationality of AI’s decision-making process [6]. Is it this process of decision-making that implicates unthinkable, non-emotional, ruthless, and savage ability that renders AI technologies as more rational decision-makers than humans as per according to [2].

Reflections of AI Technologies

AI technologies have been deemed to be an important of contemporary society. The abilities of AI technologies to increase efficiency, protect the environment, and increase accuracy in decision-making processes have been regarded as essential within today’s societies [3]. Despite such, AI technologies do possess certain downfalls. Notwithstanding AI’s valiant efforts in increasing efficiency, decreasing environmental degradation, and reducing the vicious capitalistic hold (i.e., what can be considered as human slavery), AI may come to eventually replace humans in certain jobs which may require actual human contact at the end of the day.

That said, the omnipresence presence of AI technologies possesses the potential to lead to a sort-of technological domination over humanity and perhaps in a grandiose discourse, eradicate any further evolutionary steps for humans wherein specific jobs are concerned (for example: surgeons, factory workers, customer service workers). Meaning, despite the abilities of AI technologies to provide superior customer service over humans as well as the capabilities of AI technologies to perform certain tasks more efficiently and accurately than humans, this technological revolution can possibly increase the capitalistic hold over society via AI technologies faster and more precise production of goods and services whereby more product is manufactured with less sells in retails due to the replacement of humans and therefore creating a loss of jobs and wages. Grandiose thought indeed however, when reflecting on what is going on in the world in current day, such as robotic surveillance dogs during the COVID-19 pandemic in China, robotic servers in various countries, surgical technologies promoting the use of robotic technologies during real-life human surgery, self-checkout kiosks, 24/7 depanneurs that do not have any human staff present; in addition to various movies and another Netflix show – Better Than Us, it is clearly observable how AI technologies have already begun to reduce the presence of humans in certain jobs.

[7] stipulates in his coined term surveillance capitalism, such aims to dictate a “…new form of information capitalism…[purposed] to predict and modify human behavior as a means to produce revenue and market control” (p.75). That said, the dialect mentioned above holds a certain truth and reality within contemporary society today whereby AI labour provides a new avenue of thought to re-image how certain human-lead jobs are now being robotically transformed; that-is-to-say “…replace the human body with machines that enable more continuity and control” as [7] suggests. With that said, AI technologies possess the potential to revolutionize and feasibly even innovate spaces that humans did not think was even possible (such as major intricate surgeries for example). AI technologies in this sense can be argued to therefore hold both positive and negative attributes towards their contributions in modern-day contemporary societies today.

AI Technologies and Therapy - Analyzing and Evaluating the Impact And Utility Of AI Technologies Within Emotional Management Therapeutic Attempts

Robot Therapy

Robot therapy is observed to be on the rise world-wide. Robot therapy can take the two form types – (1) robot-assisted therapy (which constitutes of therapy programs designed by healthcare professionals such as doctors, nurses, social workers, etc.,) and (2) robot-assisted activity (which refers to patients interacting with robots with no specific therapeutic goal associated with the interaction) [8].

Paro, a mental commitment robot (a robotic stuffed animal sort-to-speak) was constructed for therapeutic purposes and tested in pediatric wards as well as in elderly institutions in several countries. The results yielded by this experiment showcased that through interactions with Paro, patient’s moods improved, enhancing activity amongst patients as well as increased communication between other patients and their caregivers [8]. In addition to such, urine tests sampled from the patients whom interacted with Paro indicated reduced stress levels as well [8]. Paro provides concrete proof that robot therapy can have a positive impact on patient behavioral changes via AI technologies. Paro falls within the umbrella term of AI given that the devices used within this genre of therapy perform a task programmed into the mechanism.

Moreover chatbots, which are machine agents that service patients in natural language are also used by patients in modern day as an anonymous, asynchromatic, personalized, and consumerable method to seek help [9]. Chatbots have been therapeutically employed to help assist patients in effecting behavioral changes amongst patients who suffer from a myriad of different diagnosis including mental health/neurological disorders, addictions, sexually-transmitted diseases, nutritional-metabolic disorders, mental and physical wellness and others [9]. Chatbots are explained to have been designed to help and assist patients in their decision-making processes by counseling patient’s awareness of their mental state; providing goals and solutions towards patient problems [9]. Chatbots have proven successful in altering patient behaviors through the bot’s abilities to foster a secure and anonymous setting that would typically make patients feel less inclined to speaking to a human therapist based on emotions of fear, shame, and embarrassment towards the problem that the patient is facing [9]. Not only so, chatbots can also gage the patient’s mood and affect change via the abilities of chatbots to capture patients’ facial expressions through a camera, or analyzing text messages, which enables the chatbot to algorithmically respond to the patient in the most appropriate manner (e.g. responses directed towards attitudinal changes, jokes, motivational interviewing, etc.) [9]. Just as a human therapist can personalize a rehabilitation/treating plan for their patients, chatbots can do the exact same based on the information provided and gathered throughout the chatbots and patients therapy sessions [9].

Importantly, the study conducted by [9] refers to the fact that patients are more inclined within today’s contemporary modern-society to seek robot therapy due to factors associated with time, cost, anonymity, and secure therapy sessions based on decreased sentiments of embarrassment and shame that would otherwise be felt when confronting a real-life human being capable of feeling the same emotions as they are. The idea of trusting and communicating with a robot that does not have real-life friends to talk to and vent about their day as a human therapist would, or cast judgements towards the patient, presents an appealing attraction towards robot therapy as data amongst study participants [9]. In conjunction with such, [10] found similar results in their study concluding engendered data that described their participants acceptance towards the use of social robots as a source of a more economically friendly therapy option, more available/accessible, neutral, and anonymous therapy option; as well as a more interactive and affirming option of therapy, a more personalized form of therapy that holds the ability to be more open during sessions (p. 1083-1087).

Similarly, [11] found that automated telehealth online services that provide web-based educational programs for individuals suffering from anxiety and depression proved successful in improving depressive symptomologies amongst the patients using uniquely web-based lessons with no therapist present (p.83). The researchers also found that telehealth mental health programs not only provided a cost-effective therapy option, but also helped to reduce perceived and felt stigmatizations associated with seeking professional mental health therapy [11].

Telehealth

Telehealth throughout the years has proven to be a time-effective, cost-effective, and accessibility-effective solution to seeking therapy from professionals or web-based applications developed by healthcare professionals. Multiple phone applications throughout the past decade have emerged such as MoodKit, Mind Shift, Head Space, Calm, Unplug, and many others; all with the objective to help patients find and access mental health care. Telehealth used by doctors, nurses, social workers, etc., have also been increasingly utilized in order to better service patients; especially during the COVID-19 pandemic. Telehealth has provided an innovated avenue for patients to receive care and services from professionals in reaching more rural communities and other populations which would otherwise encounter more difficulties in accessing care and services. Telehealth also offers the possibility for patients that are more hesitant towards seeking mental health services to access services in a more anonymous manner (one does not need to provide one’s name, location, or even show one’s face if they do not want to), creating a more enticing environment to seek help [12].

That said, what is telehealth? Telehealth in its simplest of forms denotes healthcare that is provided by means of telecommunication [12]. Telehealth invokes the long-lived question as to whether or not a better health-care system is possible [13]. Telehealth brings with it the possibility to diagnose and provide treatment to individuals at a more efficient rate [13]. Additionally, telehealth also presents as an opportune alternative to in-person therapy for patients that cannot otherwise attend in-person sessions due to physical disabilities, time or geographic limitations [14]. [14] found that patient with lower activity levels (i.e., motivation to leave the house) actually increased when presented with the option to attend therapy sessions online, consequently increasing the positive therapeutic outcomes and lowering experiences of pain. Telehealth in this sense does require the use of an AI technology specifically, however dedicated professionals working in telehealth are found to employ AI technologies in the delivery of their services.

[15] found that the use of chatbot AI technologies are used by an array of mental health practitioners in their practices servicing patients. That-is-to-say, said practitioners can use AI technologies not only in services execution (having ChatGBT open and listening during the session to provide responses to patients for example), but also utilize AI technologies in patient note dictation, tailoring interventions, double-checking treatment plans, refining patient discourse(s) as well for example [15]. In this sense, AI technologies provide practitioners with innovated assistance in the delivery of therapy.

The use of AI technologies by practitioners raises interesting questions. For example, can AI technologies when used by practitioners during live sessions provide improved or “…appropriate recommendations given a specific user utterance and mental state[?]” [16]; Does the use of AI technologies by practitioners during live sessions encroach on patient confidentiality? [17]; Does the use of AI technologies by practitioners during live sessions perhaps help to improve the quality of services provided in supporting practitioners to better decipher through what is being explained by a patient via the use of AI technologies providing live feedback? [18]; Does the use of AI technologies by practitioners during live session improve the quality of sessions with patients through the collaborative use of technology and human? [18].

In conjunction with such, [18] in their study report interesting findings wherein AI technologies is used for emotional dysregulation therapy. The authors present data on how interactive platforms focused on providing users with immersive environments where AI technologies can help users to detect their emotional state and therefore try and regulate their feelings [18]. The study reveals how apps use “…AI-enabled biofeedback and neurofeedback systems can help individuals gain awareness and control over their physiological responses associated with emotional dysregulation… [using systems that provide feedback on] heart rate, brain activity, and other physiological indicators, allowing individuals to learn how to modulate their emotional responses.” [17]. The authors also note that in the event that the user should require real-time support, virtual therapists as well as chatbots are also available to patients consulting the telehealth technology [17].

When considering the data reported by [18], the author alludes to an alluring point when reflecting on the use of chatbots as a collaborative partnership. Meaning, the understanding that AI technologies such as chatbots pulling information from all over the Internet (i.e., from different studies interventions, approaches, various professors/researchers/theorists/philosophers/social media/etc.) to formulate a response for inputted data, such presents a curious debate wherein considering economics, appropriate content, exploitation, and information being distributed is concerned. In an era where technology is a pillar in day-to-day functioning, notions of practitioner efficiency and productivity must also be thought of. As what a chatbot can issue in terms of response may not be in-line with what the practitioner necessarily believes in or uses within their practice. Such then puts into question the quality of services provided by the practitioner and their abilities to efficiently retract unmitigated information that may be contained within the chatbot response during a live session.

When considering such, are practitioners becoming lazy or innovative? When contemplating the use of AI technologies via chatbots by practitioners during live therapy sessions such presents intriguing reflections on how the world of therapy on the side of practitioners will need to be revised wherein privacy, anonymity, confidentiality, and human ability is concerned. Considering such, practitioners using AI technologies within their services sheds light on an interesting facet regarding the use of AI technologies in the world of therapy – will technology take over the use of humans in a field of work based on emotions and human understanding? Fruit for thought.

Case Study of Netflix show “Too Hot To Handle”

The Netflix series “Too Hot To Handle” hosts multiple seasons of casted hot, single, and emotionally disconnected participants for a cash prize earned at the end of the show based on who has made the most emotional progress by following AI robot Lana’s island rules. The spectacle is purposed to showcase the harsh reality of today’s singles and the emotionless hook-up culture engaged in, in today’s societies all around the world (the Netflix series is an international phenomenon that is filmed in America, South America, and Europe). AI Lana’s objectives are not only to keep the contestants abstinent, but also to keep them from making bad choices and show them how to form meaningful romantic connections via her interventions through AI technology (i.e., a cone).

The AI device used in this Netflix show whether scripted or whether using AI chatbot technologies, has been observed to ostensibly successfully been able lead her participants to feel foreign emotional experiences (i.e., sentiments of vulnerability and connection) which led them to altering their behaviors as analyzed throughout the television series. Lana was observed to employ the ostensible approach of conditioning – providing participants with a cash-prize in exchange for learning truer emotional connections. Note, that participants do not commit to the show knowingly being sent to a sexless retreat beforehand. Lana incentivizes her participants in the Netflix competition to face their fears of commitment, love, and emotional awareness by working their weaknesses of forming meaningful connections via experimenting in real-life situations and workshops. Said series of therapeutic workshops include “vagina workshop” (girls only), vulnerability workshops, and communication workshops. It should be noted however, that the workshops are given by humans in-person with show participants. Moreover, Lana also includes an interesting practice where once it is felt that a couple has developed an intimate and meaningful connection with each other based on sincere and authentic developed a meaningful relationship, Lana provides said couple with a green light to kiss. Thus, highlighting to the participants what a meaningful connection feels, looks, and sounds like in terms of a romantic relationship.

Contemplating the Netflix show “Too Hot To Handle” in relation to chatbots, Lana whether scripted or not illustrates how both one-on-one discussions with participants as well as responding to group prompts utilizes AI chatbot technologies to help structure and guide therapeutic understanding wherein consequences, actions, and purpose of emotions are concerned. When reflecting on the premise of the show, the program seemingly presents an attempt to uncover if individuals would better respond to a robot than to an actual real-life human therapist via the comparative use of therapy approaches (i.e., Lana (AI) and in-person coaches). The show also admittedly provides insight to the effectiveness of a mixed approach (i.e., combination of AI and human intervention) in this sense. It should be noted that Lana from the Netflix series is not a truer form of AI. Meaning, Lana is evidently scripted by show writers and not a responsive technological body. Notwithstanding such, the show presents an interesting perspective on chatbots and the presence that AI technologies are beginning to have wherein the rehabilitation of emotions and feelings. In light of this creative example, AI technology drive therapeutic attempt introduces an interesting example at tackling emotional therapy via the use of AI technologies.

Discussion

[19] writes an interesting reflection on AI technologies and human intellect:

“Any machine is always a machine of cognition, a product of the human intellect and unruly component of the gears of extended cognition. Thanks to machines, the human intellect crosses new landscapes of logic in a materialistic way…)” (p.1).

AI technologies used by both patients and practitioners in this sense to the quote above, make sense of the evolution that emotional therapy in mental health is taking in light of AI. Chatbots in this sense provide an example to this. Chatbots have been elaborated as a sort of robotic technology that engages in both narrow and general AI intelligence, perpetuating effective therapy sessions for its patients by both performing latent and systematic functions, whilst also executing more complex interventions separate from human intervention. Evidently, AI devices are able to provide a rational decision within a time-efficient window. That said, it is believed that this fact alone stands as being an important differentiating factor between human and machine interventions in distributing practical advice and coaching within a timely manner. Thus, situating the use of AI technologies in emotional therapeutic approaches as a new, important, and pertinent innovation in the world of therapy and rehabilitation.

Withal, therapy is a subjective experience, in that therapy sessions provide individuals with the opportunity to vent, decipher, reflect, question, realize, cry, yell, find happiness, etc. That said, AI technologies utilized in successful therapeutic practices is dependent on the individual whom is seeking the help. However, when discussing the effectiveness of AI technologies in emotional therapy, it can be stated that AI technologies provide both patient and practitioner with certain advantages whilst also with disadvantages. AI technologies provide greater opportunities for anonymity, cost-efficiency, time-efficiency, improve interventions and care plans as well. Albeit, AI technologies used in therapeutic attempts can also present misguided information, interventions, and treatment plans should the practitioner not be verse or is efficient in the use of AI technologies within their practice. What needs to be better understood in the realm of AI technologies used in therapeutic attempts however is the use of academically trained humans versus AI technologies collecting information from the Internet in order to provide appropriate support to patients; AI technologies should be consulted with professionals whom have mastered the field and not any AI technology that takes anything and everything off of the Internet. AI technologies also need to be further elaborated in terms of privacy and informed consent of patients wherein the use of AI technologies by practitioners during live sessions is concerned.

Conclusion

When analyzing the use of AI technologies in therapy, it is plausible to consider that therapy using AI technologies provides both patients and practitioners with an evolved type of approach to emotional therapy. The concept of confiding in an AI device wherein emotions are concerned continues to be trivial within its application given the context of emotions - how the cold hard-drive of AI can be considered more appealing than a warm human. Notwithstanding such, AI technology used within therapy in today’s contemporary society is understood to provide an array of benefits not only for patients but also for practitioners as well. Availability and openness of information concretized by a chatbot in developing interventions and treatment plans, as well as providing personalized responses based on body and eye contact is revealing to a new era of creativity and innovation wherein human needs are concerned.

It would seem that patients find a comfort in chatbot responses derived from AI technology when seeking help. Does this have to do with the idea of information precision or perhaps the lack of trust in humans? AI technology used in mental health therapy showcases an interesting link between wanting to have human emotions explained yet resorting towards a technological device or platform that is supposedly not able to feel or conceive human emotions; but rather simply explain to humans what they are programmed to search for or respond with based on the set algorithm. When reflecting back to the Netflix show “Too Hot To Handle” and chatbot technologies, as [17] report, chatbots and AI devices need to be pre-programmed by a human to identify algorithms to start, whereby the AI device can then proceed on its own in identifying information to the user. [17] also emphasize that with constant technological improvements, chatbots used in therapeutic attempts will also need to be consistently refined and updated by a group of qualified professionals in order to assure not only clinical precision, but also to ensure that AI technologies aid and not replace humans; using “Too Hot To Handle” Lana as a creative example. All-in-all, this review evaluates data found in various examples investigating AI technology and their use in therapeutic attempts towards emotions as something that seems to be ostensibly preferred amongst patients in current modern-day contemporary society. Albeit, the success rate of patient rehabilitation using AI technology versus human led interventions must be better understood in comparison to each other as well.

Limitations

Limitations found within this paper include the fact that the scope of the review is limited to English-language sources, utilizes a non-clinical source such as the Netflix show “Too Hot To Handle which does not possess any real-time evidence nor purpose towards AI technology-led therapy, as well as is lacking in meta-analysis.

Future Research

Future research can use this review to assess the authenticity between human interaction and AI interactions, deciphering through the interchanging differences between what an AI robot says and what a human says base on searching for information within different database versus what is learnt in an academic facility by a human. In this sense, does the AI technology inherently present anything more profoundly different than what a trained human does and the therapeutic effects such has on the rehabilitation of the patient.

Declarations

There are no declarations to be made for this project.

Ethical Approval & Consent Statement

Not applicable – no research subjects.

Funding

No funding was received for this project.

Availability of Data and Materials

Information, referenced articles, and statistics cited throughout this review can be accessed online.