Computer Chronicles Revisited 58 — Mind Over Minors, Relax, and the Therapeutic Learning Program

“Expert” systems were all the rage in the late 1980s, as demonstrated by a previous Computer Chronicles episode on their use in law enforcement. But expert systems tended to be expensive–costing tens of thousands of dollars–and thus only available to businesses and institutions. Yet by 1986 there were efforts to market lower-end systems to individual consumers. This next Chronicles episode from February 1986 looks at the meeting of the minds, as it were, between expert systems and another mid-1980s boom market: pop psychology and the self-help movement.

Stewart Cheifet and Gary Kidall opened the program by looking at IntraCourse, a program running on an IBM PC. Cheifet said the program’s authors claimed it could help couples with “sexual compatibility or incompatibility problems.” Cheifet prepared a profile of a fictional couple, answering about 100 questions posed by the software, which then produced a series of charts and a “compatibility percentage” based on 12 different characteristics.

Cheifet noted this was just one example of the host of “psychological software” that was now out on the market. But how could a dumb, number-crunching computer do anything useful with regards to human behavior? Kildall said that computers had been used for years to tabulate scores from “objective” tests like personality inventories. When it came to non-objective programs like IntraCourse, they might be a lot of fun but there was a real question about how valuable the computer analysis was. But he said it might be possible to use “expert” or rule-based systems to get some answers.

Using Cognitive Software to Help Trauma Victims

Wendy Woods’ first report focused on the use of computers by doctors to treat patients recovering from severe head trauma. Woods said such patients typically required lengthy rehabilitation to try and regain basic cognitive functions. Exercises to improve concentration, memory, or motor skills involved repetitive games administered by the therapists. But computers were changing cognitive therapy, often with impressive results.

Woods said that mental exercises most suited to head injury patients were easily adapted to a computer program. Specialized software like a program developed by the Karen Lambert Foundation helped patients to redevelop lost cognitive skills with logic-based games. (The name of the program was apparently Memory Master.) Doctors didn’t expect computerized routines to replace therapists, who sat by patients during sessions, but in some cases patients could work by themselves. And the challenge of keeping up with the computer kept them motivated.

The program also seemed to provide patients with a tactile sense of their own progress, Woods added. As they advanced in the program, head injury victims increasingly felt their diminished capacities. So a program that could gauge their own progress was psychologically stronger.

Leading a brain-injured person back into the mainstream of life called for a combination of physical and psychological therapy, Woods said. While still in an early stage, cognitive software seemed to be making a difference. Rehabilitation centers could use their staff more efficiently, and most importantly, patients could go home sooner.

Helping Parents Prepare Personality Reports on Their Children

Dr. Jim Johnson and Dr. James Gardner joined Cheifet and Kildall in the studio. Johnson was a psychologist and president of Human Edge Software. Gardner was a psychologist with the Fairview Developmental Center in Orange County, California.

Kildall asked Johnson if it was really possible to use a computer program to get meaningful results from a subjective analysis of someone’s behavior. Johnson said you could do a surface-level analysis with a computer. It couldn’t do the same thing as a professional. But you could do more in an interactive mode as opposed to a book.

Gardner said he was more cautious. His center also used computers, but they spent a great deal of time assessing the validity and accuracy of the data before it was entered. And they didn’t run a program unless that information was reliable; if it was not, they required people to go back and change the inputs until they reached a certain level of accuracy.

Kildall noted that the software Gardner referred to was used by professionals, while Johnson’s programs were marketed towards consumers. Kildall asked Johnson for a demonstration of one of his programs, Mind Over Minors. Johnson explained that with Mind, a parent assessed themselves and their child, and then the program gave the parent a report. The idea was to categorize both parties and then “tailor down the strategy” for how the parent could better interact with a child.

Mind presented the parent with a list of roughly 50 adjectives to describe them and their child, such as “anxious,” “bossy,” and “active.” The parent then entered whether or not they “agreed” or “disagreed” that the adjective described the relevant person. Johnson prepared sample profiles for his demo. He then showed the final “Strategy Report,” which also referred to a book that accompanied the program. The first part of the report offered a general description of the parent and child based on the responses to the various adjectives. Then it provided specific advice to parents regarding techniques they could use with their children. Johnson claimed that the software had “boiled down the expert literature” related to parent-child interactions and only offered the advice that was relevant.

Kildall asked if the technology behind Mind was based on a more professional system. Johnson said yes, there were about 5-to-10 man-years put into Mind. (Johnson lamented it took a lot longer than he thought it would.) To build the program they had to go through a lot of expert books and research. He explained the final program was “useful to parents” but still cautioned that there was a “tongue-in-cheek attitude” with this type of software. Nevertheless, Johnson said he was a former professor and scientist, so he wouldn’t put out anything that didn’t involve a good deal of research.

Kildall asked Gardner for his opinion of Mind. Gardner said it was a valuable tool for people to deal with non-serious problems. But the kind of work he did involved people with severe aggression and self-abuse. The kind of software necessary to deal with those problems required 2,000 bits of information as opposed to 50.

Kildall followed up, asking Gardner if the software he used was widely utilized by other psychologists. Gardner said yes, it was used by people in mental health centers, group homes, schools, and state institutions. The software provided the equivalent of going to a university and getting 24 disciplines to sit down for a week and tell you what was going on with the patient.

Cheifet asked Gardner to elaborate further on the “expert system” he used. Gardner said the program dealt with about 5,000 “rules” that were extracted from the professional literature on a variety of subjects such as medication, nutrition, and physical health. The program then used an artificial intelligence module that could change the rules based on experience. So the program actually adapted itself over time to the environment where it was used.

Cheifet asked about the potential risks of people getting the wrong type of advice from these programs. Gardner said they spent about one-third of their time ensuring the information provided to the software was accurate. Indeed, the program wouldn’t even run unless it had reliable, valid, and accurate information. The reason for that was to minimize those risks.

Johnson added that over the years experts had often given conflicting advice, especially with respect to subjects like child-rearing. He said with respect to his own products, he tried to do the best he could by trying to tailor advice based on the available research as opposed to simply offering generic advice. So he saw the computer as an improvement rather than a danger.

Kildall asked Johnson what the professionals in his field thought about his software. Did they see it as a threat? Johnson said reaction was mixed. The whole issue with AI was that once you developed a system that in some way mimicked an expert, they started to feel threatened.

This Was Perhaps the Silliest Thing They Made Wendy Do for This Show

Unlike most of her remote segments, Wendy Woods took a more hands-on approach with this next report from the Biofeedback Institute in San Francisco. More precisely, she took a head-on approach, using a headband equipped with tiny sensors to “control” a computer program called Relax, developed by Synapse Software.

A woman with dark curly hair and glasses seated in a chair. She is wearing a large black headband and has a bored expression on her face.

Dr. George Fuller Von Bozzay, a psychologist and director of the Biofeedback Institute in San Francisco, used a Commodore 64 equipped with an interface box and special software (including Relax) to conduct his biofeedback work. Von Bozzay told Woods that biofeedback software could now be used with most people who had psycho-physiological disorders. For example, a patient could learn by watching the video screen and see where their heart pressure, blood rate, or muscle tension was and learn how to “vary” those meters or other graphics on the screen.

Woods explained–and attempted to demonstrate–that in one part of the Relax program, the patient tried to lower a balloon on the screen by relaxing. The program also included a kaleidoscope that increased in complexity and beauty as the patient became more at-ease. And there was also a graph that “more scientifically showed” a patient’s progress.

Woods said Von Bozzay used another program to monitor a variety of body functions visible on the screen. He said the “real key” to biofeedback was the process, not the machine. So the more the machine distracted the person from the learning procedure, the less learning that took place.

But how well did it work, Woods asked? Von Bozzay said he was pleased with his patient’s progress. And Woods herself said–perhaps facetiously–that she “looked pretty relaxed.”

The Dangers of Software Psychology

Dr. Roger Gould and Dr. Bernie Zilbergeld joined Cheifet and Kildall in-studio for the final segment. Gould was a psychiatrist and president of Interactive Health Systems in California. Zilbergeld was a psychologist and author.

Kildall asked Zilbergeld, a critic of the use of computers in psychology, about the danger of using software in place of a traditional therapist. Zilbergeld said there were quite a few possible dangers. He specifically called out Jim Johnson’s Mind Over Minors from the previous segment as raising a number of problems. Essentially, Zilbergeld said the program was a “toy” with nothing scientific behind it. He had no idea why someone would pay $50 for a program like that as opposed to buying a paperback book for $3 or $4 with the same information. If parents saw Mind as a gimmick or something to play with, that was fine. But he was concerned that some people might take it seriously. On the other hand, he said the database described by James Gardner had some interesting possibilities. We wouldn’t know for sure until it was tested, Zilbergeld said, but it certainly accounted for more variables than any human therapist could keep in mind at any one time. So he’d like to see further research in that area.

Cheifet then turned to Gould and asked him to explain his software, the Therapeutic Learning Program (TLP), which was used in group therapy settings. Gould said most people went to a therapist for short-term work because they had some problem, or symptoms that were secondary to a problem. And it was easy to identify the source of a problem: A person was in some state of dissatisfaction and uncertain how to act. Once you were sure and determined the right course of action, that dissolved the problem. So his program was organized so you could take any problem and begin to ask the question, “What might I do better or different?” And was that the right thing to do? This question had two levels: First, was it the right choice. Second, was it right for the person psychologically?

Gould said TLP was used in connection with a mental health clinic, a psychiatrist’s office, or even a wellness program. In the room there would be 10 computers, one therapist, and 10 patients. The computer asked each patient a series of questions designed to identify their specific problem or dissatisfaction. He showed a sample screen that asked the patient to select one or more items from a list of sentences starting with, “I’d like to…” For example, one of the options was, “I’d like to be on time with my appointments and other commitments.” The idea was to identify those items that addressed the patient’s specific problem.

Kildall asked what the program actually did with all of this information. Gould said the program continued to guide the patient’s thinking process. But there was no prescription or evaluation. Kildall asked about the actual results in group therapy. Gould said he’d used his program with about 1,000 patients so far, and about 75 percent of them had acted on their problem after completing the therapy program, which lasted a total of five sessions. Gould said this “alternative method” was much more effective than what therapists like him had been doing previously. He said this process was also more useful for the patient because it increased their self-esteem and rendered the role of the therapist less important.

Cheifet asked Zilbergeld for his thoughts on this. He said Gould’s work sounded interesting. But he said the outcomes had to be taken with a grain of salt. You could take any therapist from any school who could say the same thing. The real proof–and he said Gould knew this–was in comparative outcome studies. The real research had to be done by someone else, so we would not have answers until several years down the road. Without such research, Zilbergeld added, you just had a lot of wild claims. For instance, some people said computers would lower the price of therapy, but Zilbergeld said it was equally likely it would raise prices. After all, computers were not that cheap, especially if you had to buy lots of them.

Kildall asked if a really good “expert system” would be of genuine help to a therapist. Zilbergeld said he understood Gould’s program as an attempt to automate some of the therapist’s work. Nothing Gould said concerned him that much. But in broader terms, using computers didn’t make anything more scientific. Gould interjected that his system reflected about 80 years of clinical experience by three professors of psychiatry.

Could Computers Make Competent–if Not Expert–Therapists?

George Morrow’s closing commentary focused on what he called the “flagrant abuse” of the word “expert” in conjunction with expert systems. He said it was one thing for programmers to abuse the word. It was quite another for psychologists to do so. You would expect someone who understood human behavior to have more common sense.

He noted an expert was someone (or something) who understood a subject so well that their responses were automatic and based on patterns rather than rules. Experts knew that rules had to be broken some of the time. Knowing when was the difference between “expert” and “competent.” Given our limited understanding of how the human mind worked, Morrow said nobody should claim to be a genuine expert. Otherwise, therapy wouldn’t be so hit-or-miss. So in theory, a computer could do as well as a human therapist, at least when it came to helping a patient discover how to solve their own problems.

Commodore Received Loan Extension, Kaypro Axed CP/M Models

Stewart Cheifet presented this week’s “Random Access,” which he recorded in February 1986.

Human Edge’s Aborted IPO Led Johnson to Shift Focus to Advertising

This was the second Chronicles episode to feature a product from Human Edge Software. The first was back in early 1984, when Jim Chapman, a vice president at the company, demonstrated The Sales Edge, a desktop PC “expert system” designed to help salespersons. Much like Mind Over Minors, Sales Edge used responses to a series of yes-or-no questions to generate a customer profile.

Jim Johnson himself was a former salesman for IBM. In an August 1984 story for PC Magazine, Connie Winkler said Johnson attended the University of California, Berkeley, in the 1960s as a graduate student in philosophy but was “booted out” for participating in an extended series of protests known as the Free Speech Movement. Johnson and his wife had their first child soon thereafter, so he decided to get a job selling Jell-O.

Fully abandoning his counter-culture roots, Johnson later went to work for one of the great symbols of establishment power–IBM. He later told Winkler that while working for Big Blue he “was involved in a $10 million sale of a 1,000 terminal on-line system for a giant finance company.” Johnson said he used the commission earned from that sale to return to graduate school, this time in psychology at the University of Minnesota. At that time, Minnesota received a good deal of grant money “to study whether the computer could evaluate psychiatric patients and lead to better patient care,” Winkler said.

After completing his doctorate in 1972, Johnson went on to work as a professor at the University of Utah, Eastern Virginia Medical School, and the Illinois Institute of Technology. During his time at Eastern Virginia, he started his first company, Psych Systems, Inc., which built upon the work he started at Minnesota and sold computer-automated psychological tests. After taking Psych Systems public and selling his interest, Johnson used those proceeds to start Human Edge Software in 1983.

As I discussed in the previous article, Human Edge initially focused on marketing low-end expert systems to businesses. The Sales Edge was part of a series that also included The Management Edge and The Negotiation Edge. These programs sold for $250 each and could be run on a microcomputer, as opposed to more complex expert software that cost tens of thousands of dollars and required a mainframe or minicomputer. Targeting the business market proved moderately successful, at least initially, and Human Edge said it had sold about 10,000 copies of its various products by mid-1984.

Mind Over Minors was part of Human Edge’s attempt to expand its product line into the home computer market. There was also Mind Prober, a $50 program–$30 if you had a Commodore 64–marketed as something of an entry-level expert system that could help individuals generate psychological reports on friends, partners, or even acquaintances. And if that sounds creepy, well, Human Edge openly touted the “Orwellian” nature of the program as a feature, not a bug.

Mind Prober and Mind Over Minors both came with accompanying books. Indeed, it seems as if the software was simply a marketing tool to sell the books rather than the reverse. As Bernie Zilbergeld noted in his critique of Human Edge Software, a person could easily buy a self-help book for just a couple of dollars. But adding the software component–essentially a fancy version of a magazine personality quiz–let you charge $50 for the same text.

Critical reviews of Mind Over Minors ranged from tepid to mildly positive. A November 1986 article in the Wilkes-Barre Citizens’ Voice said the program was “incredibly accurate and helpful” in creating a profile for the reviewer’s daughter. Yet Dan Gutman, writing for the McNaught Syndicate, found that when he entered information about his two young nephews, the software “missed entirely” with one of them. Gutman was also “disappointed” with the overall lack of insight in the computer-generated reports, noting they offered advice that was no different than most child-rearing books. (Gutman did, however, praise the accompanying Mind Over Minors book.) And as was discussed in the episode, Gutman said the “program’s accuracy depends entirely on the parent’s assessment of his or her own personality and that of the child.”

Mind Over Minors was likely the last major product released by Human Edge Software. In July 1985, Human Edge Software filed a registration statement with the SEC, the first step towards making a public stock offering. The statement proposed a $5.2 million offering, selling 833,750 shares at $6.25 each. But before the IPO took place, the SEC’s auditors found that Human Edge was–to put it politely–massaging the books.

In a September 1987 statement, the SEC accused Human Edge of failing to “provide adequately for sales returns and/or bad debts and recorded sales in transactions,” as well as “misreporting” prepaid expenses. Basically, I think what the SEC said was that Human Edge wasn’t properly accounting for returns of its software and overstated its research-and-development costs. In financial terms, the SEC said Human Edge reported a $7,000 profit in 1984 when it should have reported a $111,000 loss.

Human Edge reached a settlement with the SEC. The company admitted no fault but did not challenge the suspension of its registration statement. This effectively canceled the public offering, and Human Edge Software was out of business by the end of 1987.

After Human Edge’s demise, Johnson shifted from the psychological software business into marketing. He started Connected Brands, a business-to-business advertising agency, in 1989. Johnson served as that company’s CEO until 2001, when the company merged into a Delaware corporation of the same name. (My assumption is that Johnson sold the business.) In 2002, Johnson founded his last business, Home Care Assistance, which provides in-home care for senior citizens living in northern California. Johnson retired in 2015. He passed away in December 2018 at the age of 78.

“Relax” Developer Another Victim in Jack Tramiel’s Battle Against Atari Developers

Synapse Software Corporation was the company behind Relax, the subject of Wendy Woods’ questionable demonstration in this episode, as well as Wizard of Wall Street, which received its third Chronicles mention in this episode. Paul Schindler previously reviewed Wizard, and Stewart Cheifet and Gary Kildall demonstrated it during the opening segment of the previous episode on investment software.

Sadly, Synapse itself no longer existed in early 1986 when this episode aired. Its rise and fall can largely be tied to that of another company, Atari, which as I’ve documented in prior posts had its own tumultuous history. To briefly recap, Warner Communications acquired the original Atari, Inc., in 1976. During Warner’s ownership, Atari launched its flagship Atari 2600 video game console as well as a line of 8-bit home computers. When the video game market collapsed in early 1983, however, Warner broke up Atari and sold most of its home computer and video game console assets to former Commodore boss Jack Tramiel, who formed the new Atari Corporation.

Synapse entered this story in 1981. Ihor Wolosenko and Ken Grant co-founded Synapse Software Corporation that year. Appropriate for this episode’s subject, Wolosenko earned his undergraduate degree in psychology, although his first professional endeavor was running a photography studio in Boston. After selling that business and moving to California, Woloshenko told Kay Savetz of Antic: The Atari 8-Bit Podcast in 2014 that he’d started hosting hypnosis workshops and needed a way to keep track of his mailing lists. So he and Grant developed Filemanager 800, a database program for the Atari 8-bit computer line, which became Synapse Software’s first commercial product.

Synapse quickly expanded its catalog to include other programs, mostly computer games targeting the Atari platform. By April 1983, Synapse had outgrown its original “office” in Woloshenko’s Berkeley apartment and moved into a 22,000 square-foot facility housing 35 employees, according to a contemporary report in Antic magazine. In August 1983, Synapse projected sales of around $20 million and had nearly 100 people on the payroll, according to the Berkeley Gazette.

While Synapse was soaring, Atari’s collapse was in full swing. In July 1983, Warner ousted Ray Kassar, its own handpicked CEO at Atari, and scrambled to replace him before settling on James Morgan, an outside executive from the tobacco industry. At first, this was actually good news for Synapse. At the 1984 Winter Consumer Electronics Show, the Morgan-led Atari announced plans to purchase a suite of office programs from Synapse: SynFile+, the successor to Filemanager 800; SynCalc, a spreadsheet; and SynTrend, a statistics and graphing program. Atari planned to publish and sell each program for $100.

But everything fell apart for Synapse after Warner dumped Atari’s home computer assets onto Jack Tramiel. The new Atari Corporation had no intention of honoring its contractual obligations to purchase Synapse’s software. Wolosenko told Kay Savetz that Sam Tramiel, Jack’s son, literally ripped up the contract in front of him and dared Wolosenko to sue him. Wolosenko did just that, and he told Savetz that the Tramiels eventually paid up just before a civil trial was scheduled to begin.

But by that point, the damage was already done. Faced with Atari Corporation’s refusal to pay for the approximately $1.5 million in software it had ordered under Morgan, Synapse needed money just to stay afloat. In late 1984, Brøderbund acquired a majority interest in Synapse. Although Synapse remained a separate entity for a few more months, by January 1985 it had been fully absorbed into Brøderbund. Synapse continued as a separate division for about another year to develop new products. But it was Brøderbund that published and marketed Relax, Wizard of Wall Street, and the various Syn programs originally sold to Atari.

Dr. Bernie Zilbergeld (1939 - 2002)

Dr. Bernie Zilbergeld died in June 2002 due to complications from diabetes, according to an obituary in the New York Times. Although his own work wasn’t discussed in detail during his Chronicles appearance, Zilbergeld was an expert in the psychology of male sexuality. He also gained prominence as a critic of the value of long-term therapy.

Zilbergeld earned his doctorate in clinical psychology from Berkeley in 1971. He then spent several years as co-director of clinical training for the Human Sexuality Program at the University of California at San Francisco. Zilbergeld’s work at San Francisco led to the publication of his 1978 book, [Male Sexuality: A Guide to Sexual Fulfillment]https://archive.org/details/malesexualitygui0000zilb).

Zilbergeld told the Fort Worth Star-Telegram in a March 1978 interview that his book was designed to confront many of the “myths” and cultural taboos surrounding male sexuality:

There is tremendous talk among men about sex, but it is not talk that conveys information. It’s more the kind of sexual talk that shows that we belong to the same club, the locker room talk, but as far as talking about how we got introduced to sex or about any problems, men don’t do that.

Men assume there are certain standards they have to live up to, and right away they compare things to the only model they know (themselves). They don’t realize that other men have the same reactions and feelings. In sex, you learn a lot of stuff early that is of no use.

The Times obituary noted that Zilbergeld was among the first authors to publish a popular book on this subject. Male Sexuality went on to sell over one million copies and was still in print at the time of Zilbergeld’s death.

Zilbergeld wrote another book in 1983, The Shrinking of America: The Myths of Psychological Change, which addressed pop psychology and psychotherapy. In a September 1983 interview with Cheryl Lavin for the Palm Beach Daily News, Zilbergeld explained that therapy was only realistic for attaining “small goals,” i.e., to make a “modest change in a specific area” of the patient’s life. For instance, he did not consider therapy useful for “full-blown psychoses” that required medication, or even issues related to something like alcohol addiction, where a person would likely be better served attending an AA meeting.

Notes from the Random Access File