Politics by Trial and Error
Do politicians' behaviors abide by the basic laws of learning? Perhaps so.
Posted August 4, 2018
In a revealing interview shortly after Barack Obama’s 2008 election, businessman Donald Trump offered some surprisingly supportive words for the new president. Speaking to Larry King (and reported by CNN.com on February 5, 2009, Trump: Obama ‘absolutely right’ on executive pay cap), Trump agreed with most of Obama’s initial efforts to contend with the calamitous economic conditions then facing the nation. King concluded the interview by asking Trump, “If you had the power, … what would you do?”
Trump replied in his distinctive vernacular: “The new president is trying. Bush left him with a mess, a total mess in many different ways. I really think he’s doing the best he can, but it is trial and error. They try something, if it works, great. If it doesn’t work … you don’t really know … for quite some time. If it’s really wrong … we’re going to really have a mess in two years.”
Fast forward to the 2016 presidential campaign. In the April 6, 2016 New York Times’ Editor’s Blog, Anna North commented on what she called Donald Trump’s Trial-and-Error Campaign: “Mr. Trump really has no guiding principle other than his desire to win. He doesn’t think through his positions before he opens his mouth—instead, he makes comments seemingly at random, backs off if he gets criticized, then backs off on his backing off if he happens to hear some praise. On abortion as on so many crucial issues that affect Americans’ lives, he’s just spitballing.”
Nothing changed after Trump’s 2016 election. Reflecting on his administration’s rapidly shifting positions in Middle East diplomacy, Layelle Saad asked in the Gulf News on March 5, 2017: “Are the latest developments carefully calculated moves or just a haphazard testing of the waters?” She titled her piece: Under Trump, a trial and error policy emerges for the Middle East.
If trial-and-error is truly Trump’s modus operandi, then insights into his behavior might be gleaned from the science of learning. Edward Thorndike discovered so-called instrumental or operant conditioning. After studying cats learning to escape from a puzzle box by pressing a pedal to receive food reinforcement, Thorndike proposed his famous Law of Effect: behavior followed by a reinforcer is automatically strengthened, whereas behavior followed by a punisher is automatically weakened. Those reinforcers and punishers act on initially random behaviors, strengthening some and weakening others—a positively Darwinian process. His successor, B. F. Skinner, provided further key observations and analyses from his detailed studies of rats and pigeons trained in a specially-built laboratory apparatus—the Skinner box.
Now, we’re set to appreciate the latest commentary on one of Trump’s most controversial behaviors—tweeting. This analysis comes from Maureen Dowd’s July 7, 2018 op-ed in the New York Times, "For Whom the Trump Trolls." Deeming Trump to be addicted to tweeting, Dowd suggests that “the Oval Office is becoming a Skinner box” and laments that “Twitter offers positive reinforcement for negativity. [Trump’s] tweets propel the story on cable news and shape the narrative for reporters—who are addicted to the First Addict.”
Setting aside the question of whether clinical addiction is involved in either Trump’s tweeting or the media’s reporting, we do have some striking data from the Toronto Star (July 14, 2018) which speak strongly to the role of positive reinforcement in Trump’s tweeting and his other public pronouncements. Here’s the headline: "Trump has said 1,340,330 words as president. They’re getting more dishonest, a Star study shows." Reproduced below is the Toronto Star graphic which documents the clearly rising trend of false claims Trump has spoken since being inaugurated on January 23, 2017 until July 1, 2018.
This progressive rise in Trump’s false claims appears to be a textbook case of positive reinforcement. But, what’s reinforcing his behavior? Dowd suggests that those who report on Trump—often in highly disapproving ways—may themselves be responsible for this result.
Wait, you exclaim! Shouldn’t all that negative coverage punish Trump’s making such flagrantly false statements? Yes, it should—if Trump found those negative appraisals to be aversive. However, to someone who finds the limelight intoxicating, any coverage—positive or negative—can be positively reinforcing. That means that assiduously checking every one of his claims and excoriating Trump for these falsehoods—both large and small—are actually counterproductive! If that’s so, then we’re stuck in a feedback loop that’s only making the matter worse, not better.
Is there any way out of this vicious cycle? Perhaps. At least two techniques—both derived from learning science—suggest themselves.
First, reinforced behaviors generally decline when they are no longer reinforced—experimental extinction. So, the suggestion is to lower the intensity of the criticisms and to reduce the amount of coverage allocated to fact checking. This tactic will deprive Trump of what he finds most reinforcing—any form of attention. Of course, doing so will require considerable restraint; but, this strategy might be worth a try. Indeed, the experiment is currently underway at ABC News: Justin Fischel now authors Fact Check Friday, which highlights only Trump’s most consequential falsehoods of the past week.
Second, the press might positively reinforce Trump’s factual and verifiable statements. Although truthful statements may now be occurring at a low rate, they might nonetheless be encouraged by approbation. Tilting the balance of his rhetoric from dishonesty toward honesty would surely be a good thing for all concerned. It might also help undermine Trump’s potentially dangerous claim that the press is the ‘enemy of the people.’
One thing is clear: The current state of affairs is unacceptable. No president can gain the trust of the nation by engaging in ever-increasing dishonesty. The fate of our democracy may be at stake.