Should Obviousness in Patent Cases be Bifurcated to Avoid Hindsight Bias? (Part 1 of 2)

In one of the first studies of its kind, Professor Gregory Mandel from the University of Albany published an empirical analysis on the effects of hindsight bias in patent law, and made a number of surprising findings. Specifically, Professor Mandel found that

(1) the hindsight bias distorts patent decisions far more than anticipated, and to a greater extent than other legal judgments;

(2) neither the Federal Circuit's "suggestion" test nor the Supreme Court's Graham requirements solve the hindsight problem;

(3) the admission of secondary consideration evidence does not cure the hindsight bias;

(4) jury instructions that explicitly identify and warn against the hindsight bias do not ameliorate its impact; and

(5) the hindsight problem pervades patent law to an extent not previously recognized - it biases decisions under the doctrine of equivalents, claim
construction, the on-sale bar, and enablement.

An interesting aspect of the paper was that the approach used by the author was based more on a psychological analysis of subjects under test conditions, rather than relying solely on case law to divine possible instances of hindsight. One of the problems that the research identified was that a non-obvious validity requirement requires a decision-maker to first make a determination based on ex post information (i.e., that the invention was achieved) before making an ex ante judgment (i.e., whether or not the invention was obvious at the time it was made). Since the large majority of obviousness determinations start from the point of a known outcome - the invention itself - and then attempt to reconstruct the state of the art back to the time the invention was made, the effect on the qualitative judgment of obviousness becomes inherently skewed by hindsight reasoning.

A series of tests were conducted on mock jurors, where fictitious inventions were presented under various controlled conditions to see what effects hindsight had on the mock jury's conclusions. The test affirmed the findings of prior hindsight bias research revealing that individuals tend overestimate both the likelihood of a known outcome occurring, and the foreseeability of the outcome when judging obviousness:

Participants who were not informed of the invention were substantially more likely to judge a solution non-obvious than participants who were informed what the invention was. The magnitude of the hindsight bias in these patent scenarios was striking, and is greater than that reported for other legal judgments. Ex post knowledge of invention deeply affected participants' conclusions regarding whether an invention was non-obvious ex ante.

Equally significant, debiasing instructions based on actual model jury instructions did not ameliorate the hindsight bias. This finding is consistent with a study of the hindsight bias in tort law, which found no significant effect from hindsight debiasing instructions. It is also consistent with a variety of research on jury instruction indicating that the instructions often fail to produce their desired results, that jurors have an extremely low level of recall and comprehension of instructions, and that intended improvements to jury instructions often actually reduce juror understanding.

In addition to the hindsight bias, the study indicates that hindsight condition participants who concluded that an invention was obvious are more confident in their answer than foresight participants.

The study also found that jurors were just as susceptible to hindsight bias regardless of whether they were answering individually, or answering as a collective jury. Judges were also found equally prone to hindsight bias - while judges may be more generally versed in patent proceedings, they are likely not any more "familiar" with prior art non-obviousness determinations than juries are (testimony before the House Subcommittee on Courts alleged that the average district court hears only one patent trial every seven years).

And what about PTO examiners? Well, they fared slightly better (.2 standard deviation units) than judges or juries, but were still significantly prone to hindsight bias.

In sum, the results of the experiment reported here and the existing empirical data indicate that non-obvious decisions, whether by judge, jury, or PTO examiner, are routinely subject to a significant hindsight bias. As a result, patent validity decisions may be both unjust and inefficient. The decisions are unjust to the extent statutorily deserving inventors are denied patents or patent scope because the hindsight bias makes non-obvious inventions appear obvious. Validity decisions are inefficient to the extent the improper rejection of patents due to the hindsight bias reduces inventors' incentive to innovate below the socially optimal level. These conclusions, if unresolved, have powerful implications for patent law, innovation, and
competition, and therefore for the economy and society as a whole.