When examining an invention patent application, the examiner first determines whether the claimed invention is disclosed by prior art. If the invention is not disclosed, it possesses novelty. Subsequently, the examiner evaluates whether the invention, despite differences from the prior art, could have been easily accomplished by a person having ordinary skill in the art (PHOSITA) to determine non-obviousness (i.e., inventive step). However, while an invention is created in view of only the prior art, the examiner conducts the examination while having access to both the prior art and the content of the application, and this difference in perspective easily leads to divergent views regarding whether claimed inventions are non-obvious.
To address this issue in Taiwan, the Patent Examination Guidelines aim to prevent patent examiners and judicial authorities from retroactively re-evaluating technical problems and their solutions through the lens of the disclosed invention, thereby underestimating the extent of its progress and innovation. This article explores the application of the ‘hindsight bias’ principle and its impact on patent examination and judicial proceedings, based on relevant judgments from the Intellectual Property and Commercial Court, the Supreme Administrative Court, and the Supreme Court.
Concept and jurisprudential basis of hindsight bias
‘Hindsight bias’ refers to the tendency of examiners or judges, once aware of the technical content of an invention in a patent application, to unconsciously view the invention as obvious, thereby devaluing its innovation. Taiwan’s patent law emphasises that the determination of non-obviousness should be based on the state of the art at the time of filing, rather than using hindsight to consider a solution as self-evident.
Regarding non-obviousness, Article 22, paragraph 2 of the Patent Act stipulates: “Even if an invention does not fall under any of the circumstances in the preceding paragraph, it shall not be granted an invention patent if it could have been easily accomplished by a PHOSITA based on the prior art before the date of filing.” Therefore, the level of skill of the PHOSITA, the level of prior art, and whether the invention could have been “easily accomplished” are frequent primary points of contention between examiners and applicants.
The Patent Examination Guidelines specify that the examination of non-obviousness must not be based on “hindsight” derived from the specification, claims, and drawings. Instead, the invention as a whole should be compared with relevant prior art, and an objective judgement should be made from the perspective of a PHOSITA, considering the common knowledge at the time of filing.
Regarding the “motivation to combine multiple references”, the guidelines state that examiners should consider the relevance or commonality between the multiple references themselves rather than between the references and the claimed invention, to avoid hindsight bias.
Practical insights from court judgments
According to Supreme Court Civil Judgment 113 Tai Shang Zi No. 459, the determination of non-obviousness should not involve simply dismantling the individual elements or steps of an invention and then mechanically comparing them with prior art references. The focus is not merely whether execution could theoretically be successful but whether there was an incentive, a concrete factual basis, or encouragement in the specific case to prompt a skilled person to pursue and succeed in the R&D. This approach avoids a mechanical piecing together of prior art references that may lead to hindsight.
In this regard, Judgment 112 Xing Zhuan Su Zi No. 70 of the Intellectual Property and Commercial Court further pointed out that when determining whether a PHOSITA has a motivation to combine multiple references, the relevance or commonality of the technical contents of those references should be considered, rather than the relevance between the references and the claimed invention. Generally, factors such as the following should be comprehensively considered:
Relevance of technical fields;
Commonality of problems to be solved;
Commonality of functions or operations; and
Teachings or suggestions.
The relevance of technical fields is assessed by whether the technical fields of the references are identical or related, which may be evaluated based on the objects, principles, mechanisms, or functions of the applied technology.
Furthermore, Supreme Administrative Court Judgment 112 Shang Zi No. 490 states that as non-obviousness examination often involves combining multiple references, to avoid arbitrary piecing together and hindsight bias, examiners should consider whether a PHOSITA would have had the motivation to combine those references. If such motivation exists, it may negate non-obviousness. In this regard, the court considered the above four factors.
Additionally, Supreme Administrative Court Judgment 110 Shang Zi No. 597 specifies that in invalidation cases, the scope and content of prior art (evidence) should be determined first, followed by identifying differences between the invention and the evidence. Finally, it must be determined whether a PHOSITA could have easily accomplished the invention based on the prior art and common knowledge at the time of filing. The examination must not use ‘hindsight’ derived from the specification to conclude obviousness; instead, an objective comparison must be made.
Based on the court judgments above, the criteria are consistent with the Patent Examination Guidelines, which require a comprehensive evaluation of “relevance of technical fields”, “commonality of problems to be solved”, “commonality of functions or operations”, and “teachings or suggestions” in determining the motivation to combine references.
Furthermore, non-obviousness should not be assessed through ‘hindsight’ derived from the specification but through an objective comparison of the invention as a whole with the prior art from a PHOSITA’s perspective.
The absence of a quantitative standard for weighing these factors allows for a certain degree of administrative discretion. However, applicants may still challenge the examiner’s position by arguing based on these four aspects.
Practical implications
The identification of hindsight bias plays a vital role in patent examination and judicial practice to ensure that the judgement of non-obviousness is objective and reasonable. Preventing hindsight is an indispensable part of the process. Applicants, examiners, and parties in invalidation proceedings should therefore have a profound understanding of the underlying principles to effectively navigate patent practice.
