TOEIC Link Reading — Inference Bridging and Implicit Connection Recovery: How Recovering Unstated Links Moves the Reading Band from 24 to 28

Bridging inference is the reading skill that recovers the connections a business text deliberately leaves unstated. A well-written email does not spell out every logical step: it states that the shipment was delayed, then states that the client meeting has been moved, and trusts the reader to supply the causal link between the two. A TOEIC Link candidate who reads only the stated propositions — treating each sentence as an isolated fact — misses the connective tissue that the inference questions in Part Seven are built on. The candidate who actively bridges recovers the unstated link and answers the "What is implied" and "Why did the writer..." questions from structure rather than guesswork.

The rubric does not name "bridging inference" as a scoring criterion, but it sits under the reading comprehension weight that Part Seven inference items assess directly. A candidate who reliably recovers the implicit connection across a two-passage or triple-passage set typically gains a two-to-four-point band lift over the same candidate reading at the literal-proposition tier, because the inference items are precisely the ones that separate the high band from the middle band.

For broader context on connection tracking, see the reading discourse coherence and bridging inference recognition guide, the reading coreference chain resolution and entity tracking guide, and the reading anaphoric and cataphoric reference resolution across text distance guide.

Why bridging sits at the inference tier

Part Seven rewards two kinds of reading. Locating reading finds a stated fact and matches it to an answer choice — the answer is on the page. Bridging reading finds two stated facts that the writer did not connect and recovers the connection the writer assumed the reader would supply. The inference questions are engineered so that no single sentence contains the answer; the answer lives in the gap between sentences, and the candidate must build the bridge to reach it.

A candidate who reads at the locating tier answers the detail questions correctly but stalls on the inference questions, because there is no sentence to point to. The stall is what caps the reading band in the middle range. A candidate who reads at the bridging tier treats every adjacent pair of sentences as a potential inference target and asks, at each transition, what unstated step makes the two coherent.

The four bridging types Part Seven tests

Causal bridging. Two facts are stated; the cause-effect link between them is not. "The vendor raised its rates. We are soliciting alternative quotes." The bridge is because the vendor raised its rates, the company is now shopping for a cheaper supplier — and the inference question asks why the company is soliciting quotes.

Referential bridging. A pronoun or definite noun phrase points to an antecedent several sentences back, or in a different passage of a multi-passage set. "The revised policy takes effect Monday. All staff must complete it beforehand." The bridge connects "it" to the training implied by "the revised policy," not to the policy itself.

Elaborative bridging. A second sentence adds detail that only makes sense once the reader supplies the implied relationship — an example, a consequence, a restatement. The reader must decide which relation the writer intended.

Intentional bridging. The reader recovers the writer's unstated purpose. A closing line — "Please let me know by Thursday" — implies a deadline pressure the email never names, and the inference question asks what the writer wants the reader to do.

The six failure modes

1. Literal capture — answering only from stated propositions and skipping the inference items as unanswerable.
2. Over-bridging — inventing a connection the text does not license, supplying an assumption from world knowledge rather than from the passage.
3. Wrong-antecedent bridging — connecting a pronoun to the nearest noun rather than the semantically correct one.
4. Single-passage tunnel vision — failing to bridge across passages in a triple-passage set, where the answer requires combining a fact from passage one with a fact from passage three.
5. Causal-direction reversal — recovering the link but reversing cause and effect.
6. Bridging under time collapse — abandoning bridging entirely in the final minutes and reverting to keyword matching, which the distractors are designed to punish.

The four-week protocol

Week one — bridge annotation. On untimed Part Seven passages, mark every sentence transition where the writer left a connection unstated and write the bridge in one phrase. Build the habit of seeing gaps.

Week two — bridge-type labeling. Classify each gap as causal, referential, elaborative, or intentional. Naming the type sharpens the recovery and predicts what the inference question will ask.

Week three — cross-passage bridging. Move to double- and triple-passage sets and drill the bridges that span passages, the highest-value and most-missed inference targets.

Week four — timed bridging. Reintroduce the clock and hold the bridging discipline under time pressure, resisting the collapse back to keyword matching in the final minutes.

The one-sentence discipline

At every sentence transition, ask what unstated step makes the two sentences coherent — and answer the inference question from that recovered bridge, never from a single sentence in isolation. That single move is what lifts the reading band from the locating tier into the inference tier.